Reading-Valuaton Methods

Page 1 of 93
Chapter 5
Valuing Commercial Real Estate
“…Value is nothing inherent in (assets), nor an independent thing existing by itself…it is a
judgment made by individuals…”
— Carl Menger
Chapter Objectives
At the end of this chapter, the reader will be able to understand:

❖ The distinction between investment value and market value, as well as how to apply it in
real estate investment decisions.
❖ The valuation process, as well as the three methods for estimating the value of real estate.
❖ The brass tacks of real estate valuation: the elements of a cash flow projection proforma
for a real estate investment.
❖ Valuation exercises that recognise the probabilistic nature of real estate valuation.
What is Value?
“Value” is a complex, philosophical construct. Much has been debated about the notion of value1
among ancient philosophers: Socrates, Plato, and Aristotle, among others. We shall trim our
discussion down to “economic value”, and even more precisely, the “financial value” of an asset,
for a simple reason that finance takes a relatively parochial view on value, making it easier for us
to describe it. Georg Simmel, the famous German sociologist from the late nineteenth century,
suggested that value, mentioned in terms of money, facilitates the transaction of assets2. This
somewhat settles the debate for us: We can value an asset in terms of money. Some assets will sell
above their value, some below it. On average, we should expect the assets to sell at a price that
equals their value.
So, if in a sale transaction if one party makes a profit, the other must run into a loss: a zero-
sum phenomenon. But is not “one dog’s meat another dog’s poison?” In other words, the same
asset could mean different levels of value (i.e. different amounts of money) to different individuals.
1
Socrates would see value in pleasure only if it was accompanied with intelligence: “the only reason why the pleasures
of food and drink and sex seem to be evil is that they result in pain and deprive us of future pleasures.” In his famous
novel “Gaban,” Munshi Premchand takes a somewhat similar view: “Pain” can be valuable if it is followed by a
promise of better times. Thus, if extreme pain has a causal association with the following pleasure, does not it have
an intrinsic utility? Philosophy is complicated, isn’t it? Thankfully, finance is simpler.
2
https://plato.stanford.edu/entries/money-finance/
Page 1 of 93 ©Prashant Das 2021. Do not distribute.

Page 2 of 93
Many modern economists agree that something has intrinsic value if it leads to some kind
of “pleasure” or “utility.” However, Carl Menger, the founder of the Austrian school of economics,
rejected the idea that assets could provide fixed units of utility; rather, they are valuable “because
they serve various uses whose importance differs.” Menger’s “Subjective Theory of Value”
rejected the earlier concepts of value proposed by Adam Smith, wherein the value of an asset
depends on how much it costs to produce it. Menger argued that both sides (buyer and seller) may
benefit from a transaction. Consider a seller of a cinema theater who sees less value (in the theater)
than its price and ends up disposing of what he thinks is less valuable. On the other hand, the buyer
of the same theater is happy buying something that she values more than the price.
Investment value (IV) refers to a price that an investor is willing to pay for an asset,
depending on her perceived utility stemming from the investment. Market value (MV) is the price
at which an asset is being offered for sale. In this chapter, we shall primarily focus on the market
value of an asset, but also bring the investment value into the picture when needed. Buying and
selling decisions should depend on the comparison between MV and IV, as follows:
Buy when 𝐼𝑉 > 𝑀𝑉
Sell when 𝑀𝑉 > 𝐼𝑉.
Contemporary economists broadly agree that the value of an asset is a function of the following
factors:
1. Strength of the future cash flows an asset could potentially generate.
2. Credibility (or “risk”) associated with the projected cash flows; and
3. Transferability of the asset, i.e., one’s ability to sell it at a fair price when desired.
For real estate valuation, we shall interpret the first factor as our projection (“forecast”) of
future cash flows. The second factor (“risk”) will be interpreted in terms of specific valuation
metrics, such as the capitalization rate. If there is a general agreement on measuring these two
factors, the market will tend to converge towards a universally accepted valuation. However, most
valuations fail to match the transaction prices of assets. This is where the third factor comes into
play. The third factor (“popularity”) is partially explained by behavioral finance, something that
classical finance finds difficult to quantify. However, classical finance also has an explanation as
follows:
Classical financial economics assumes “perfect elasticity” in demand and supply such that the
quantity demanded (or supplied) could change within an infinite range in response to change in
prices: If prices fall too much, people will demand too much of a good; and if prices increase
excessively, suppliers will produce as many goods as possible. This is clearly is not the case in
real estate markets where the net flow of capital influences the prices too. New supply cannot be
created infinitely. The new demand is subject to the availability of capital. Reservation price
respectively refers to the maximum price a buyer is willing to pay or the minimum price a seller
expects to receive in a transaction. The reservation price depends on the investment value of a
potential buyer or seller.
Geltner (2015) provides a description of how real estate transactions work. In a market, for
an asset class (i.e., a group of assets with similar characteristics), there may be multiple buyers
and sellers with their individual reservation prices. We could plot the frequency distribution of
these prices as shown in figure 5.1. Depending on how eager the two parties are, the two curves

Page 3 of 93
may be different from each other. For example, the easier it is for buyers to raise the capital, the
cheaper it will be (i.e., smaller the discount rate) that will increase the asset price. A transaction
takes place in the price region (the shaded triangle) where the two reservation price curves
overlap. The vertical line signifies the expected transaction price3. Thus, when a buyer’s IV, as
well as the reservation price, will be higher, the (dotted) curve will shift towards the right, and so
will the transaction price of the asset. All endeavors to estimate the market value of an asset are
to identify this transaction price.
Insert Figure 5.1 here
Fair value of an asset (or liability) is the price received when selling it in an orderly transaction
between market participants at the measurement date. (IFRS)4
Market Value is the estimated amount for which an asset (or liability) should exchange on the
valuation date between a willing buyer and a willing seller in an arm’s length transaction after
proper marketing and where the parties had each acted knowledgeably, prudently, and without
compulsion. (RICS, 2012)5
After a transaction takes place, one can observe the ex-post (i.e., after the fact) price. Price
is simply the amount of money used in exchanging an asset (or good or service). We could say
that market value is an ex-ante (before the fact) estimate of the most likely price assuming a fair
market. The assumption of a fair market includes numerous conditions: both the buyers and sellers
are acting prudently in their best self-interest; neither of the two parties (buyer or seller) have
disproportionate bargaining power; the asset is listed for sale for an adequate amount of time for
price discovery to take place; the transaction is at an arm’s length6 distance, etc.
From now on, we shall use market value and value interchangeably. Value, however, is not
observable. Valuation (appraisal) is the science and art of estimating the most likely price at which
the asset will sell in a fair market.
Need for Valuation

Unlike liquid, homogeneous assets such as stocks, bonds, or derivatives, real estate assets are
heterogeneous, and unique yet pricey. While stock analysts may value a large batch of financial
assets, real estate assets must often be valued individually. The need for valuing real estate assets
can be varied. Some prominent reasons for asset valuation are listed below:
3
We call it “expected” transaction prices, as all the values (on X-axis) in the shaded region are possible in a transaction.
However, a valuer would focus on the most likely transaction price that stands on the middle.
4
International Financial Reporting Standards
5
Royal Institute of Chartered Surveyors
6
Arm’s length assumes that the two parties are independent of each other and at similar footings. For example, sister
concerns of a big conglomerate may not be considered at an arm’s length.

Page 4 of 93
Asset Transaction: Buyers and sellers need an educated opinion on the probable price of a
transaction.
Financial Reporting: Asset managers must report the value of real estate assets to their investors
even if they are not being transacted. In merger and acquisition (M&A) deals, the buying firm may
want to separate the value of real estate from the assets of the target firm.
Financing: Lending institutions need to know a real estate asset’s value before deciding on the
loan terms. When a mortgage loan is modified or transferred, both the borrower and lenders may
be interested in valuing the real estate asset.
Policy Making: Policy makers may need to assess the impact on valuation of some proposed or
implemented policy measures, such as zoning or new environmental policy.
Taxation: Real estate ownership leads to both direct and indirect taxes. On the one hand, property
taxes may be ad valorem (i.e., a proportion of the value) necessitating valuation activity. On the
other hand, depreciation expenses are a major concern for real estate owners who like to offset
their taxable income by depreciation expense. Real estate and non-real estate assets may have
different depreciation schedules. In many developed countries, real estate assets may lead to
substantial inheritance and wealth taxes. As the Indian government collected meager taxes on these
fronts but incurred substantial administrative costs related to them, the inheritance tax was
abolished in 1985 and the wealth tax was abolished in 2015. Mass appraisal of real property can
be useful in assessing the property tax rates, which may depend on a municipality’s budget
projections.
Litigation: Real estate valuation may also be needed for litigation purposes. Private landowners
may dispute the compensation from an eminent domain case arguing that the land valuation was
higher. Valuation can be powerful in litigation support, e.g., to assess the loss in property value
due to a negative externality caused by a third party.
Valuers & Regulation
CRE Valuation as a Profession

Professionals involved in the activity of valuation are called appraisers or valuers 7. Several
professionals identify themselves as full-time CRE valuers. In India, although the practice of
valuation has existed for a long time, the statutes and regulatory authorities had been fragmented.
As a result, valuation was mostly a part-time activity of professionals. Traditionally, the valuation
of real estate has been dominated by professionals with a background in architecture or civil
engineering. For residential valuation, such professionals may be adequate. Nevertheless,
valuation of commercial real estate requires real estate and finance competencies. Even finance-
oriented professionals, such as financial analysts, accountants, or consultants, must go through
7
Appraisal and valuation are usually considered synonymous. Sometimes, people consider nuanced differences
between the two. For example, while appraisal may refer to a rough estimate by anyone, valuation may imply a more
in-depth process by qualified professionals. Some people argue that appraisal is the process of assessing value and
valuation is the outcome of the appraisal process. Semantics!

Page 5 of 93
appropriate training before being able to value CRE assets. In recent decades, real estate or
financial/accounting consulting firms have come to the forefront of CRE valuation. However, with
regulatory progress in the real estate markets, we should expect the business of pure-play CRE
valuation to flourish in the coming years.
A Historical Background of Valuation Professionals

Real estate valuation standards provided by the Royal Institute of Chartered Surveyors (RICS),
originally set up in the UK (1881), offer the oldest and internationally respected valuation
standards. Beyond the UK, regulation of CRE valuation is a relatively recent phenomenon. For
example, in the US, there were no state-level regulations for real estate valuations until 1990,
although some organizations such as The Society of Real Estate Appraisers and the American
Institute of Real Estate Appraisers8 were set up as early as the 1930s. These two associations
merged in 1991 to become the Appraisal Institute.
After the infamous Savings and Loans (S&L) Crisis, the Federal Financial Institutions Reform,
Recovery, and Enforcement Act (FIRREA) was established in 1998. FIRREA provisions for the
Appraisal Qualification Board (AQB) and the Appraisal Standards Board (ASB). The AQB
administered the curriculum and exam for valuer qualification. The ASB was responsible for
maintaining and updating valuation standards. Although FIRREA was required for federal
government works only, their qualifications gained recognition by various state governments as
well. Besides, other national-level valuation professional organizations (VPO) in several countries
(e.g., the Appraisal Institute: AI in the USA) have developed their own valuation standards.
As a general trend towards standardization, the valuation practices, and ethics have leaned
towards the International Valuation Standards (IVS) set up by the International Valuation
Standards Council, 2011, London.
Regulation of Valuation Profession in India

A committee of experts (CoE)9 set up by the Ministry of Corporate Affairs (MCA) of India argued
that the government statutes have primarily focused on the consumers of valuation (e.g. focus on
questions such as what is the need for valuation, who should be hired for valuation, etc.), but the
supply side of valuation has mostly been self-regulatory.
The Institution of Surveyors (1959) and The Institution of Valuers (1968) were established to
provide guidelines to valuers. After a few failed attempts in the past, Section 247 of the Companies
Act 2013 was one of the earliest attempts where valuation was brought under an Act. The federal
government delegated the authority of regulating and standardizing the valuation practice to the
Insolvency and Bankruptcy Board of India (IBBI). In this setup, the IBBI is the principal regulator
and several registered valuer organizations (RVO) serve as front-line regulators. A qualified
individual completes requires coursework offered by a relevant RVO but must pass an examination
8
Now known as the National Association of Realtors.
9
Committee of Experts to Examine the Need for an Institutional Framework for Regulation and Development of
Valuation Professionals

Page 6 of 93
conducted by IBBI to seek registration as a valuer. The IBBI study materials prepare a candidate
for valuing two separate asset classes: (1) Land and Building, and (2) Plant and Machinery.
When this book is being written, developing valuation standards has been a rather recent and
ongoing process. In August 2019, the CoE proposed establishing the National Institute of Valuers
(NIV). As such, the valuation profession will have a similar, but independent structure as chartered
accountants.
“The process of valuation requires the valuer to make impartial judgments as to the reliance to
be given to different factual data and assumptions in arriving at a conclusion. For a valuation to
be credible it is important that those judgments can be seen to have been made in an environment
that promises transparency and minimizes the influence of any subjective factors on the process”
- International Valuation Standards Council, 2011, London
A valuer is “one who is expected to perform valuation services competently and in a manner
that is independent, impartial, and objective”
-Uniform Standards of Professional Appraisal Practice (USPAP), Appraisal Institute,
USA
Valuation Process
Despite partially being art, real estate valuation is carried out as a scientific process. The process
may be iterative and several analysts may alter the process to suit their personal style. The
following activities are involved, broadly in the sequence mentioned below:
1. Defining the problem statement: The scope of the valuation may vary. The date of
valuation is a piece of critical information. Besides, a valuer must define which components
of the subject asset are to be valued. One must categorically include (or exclude) specific
property rights from the scope of valuation. The condition of sale (e.g., foreclosure sale)
may have a profuse impact on the valuation estimate. The nature of the landholding
(freehold versus leasehold) may also make material changes to the valuation. Whether
assets such as FF&E, operating business, brand, land title, etc. are included in the scope of
the valuation may significantly alter the value estimate.
2. Collecting Information: The valuation activity relies on the quality of data and
information available. A valuer must collect information on the market conditions at local,
regional, and national levels. It may involve an analysis of macroeconomic, capital
markets, and property market information. Besides, necessary relevant information to
changes on sociological (e.g. crime), regulatory (e.g. zoning) and political fronts may
educate the valuation assumptions. A thorough survey of the property is a requirement. The
data on the subject property may be collected through a site visit, from financial statements,
and through interviewing various stakeholders (e.g. occupants, managers, local brokers,
neighbors, etc.). Property market data from the locality about sales, listings, capitalization
rates, vacancies, etc. may be collected through primary or secondary sources.

Page 7 of 93
3. Data Analysis: The quantitative and qualitative information may require processing before
being converted into valuation assumptions. After the data is thus processed, a valuer
applies different methods of valuation using the data.
4. Reconciliation: After deriving value estimates from different methods, a valuer reconciles
them into one definitive valuation estimate.
5. Scenarios: A good valuation analysis includes analysis of alternative scenarios and

resulting valuation from alternative simulations. Such an analysis provides a plausible
range within which the actual value is expected to fall.
6. Report Development: Finally, all relevant information should be consolidated in a well-

formatted and presented report. The report must specify a definitive value estimate for a
specific date, including critical considerations for valuation. A brief executive summary is
expected. The report must present all data, analysis, and exhibits relevant to the valuation.
Overall, valuation is a fascinating activity, as it allows us to convert an asset with its physical
and intangible attributes in terms of money. The valuation process requires creative analytical
skills to deal with the issue of information paucity and take the most optimal view on what the data
infers. Besides, it involves a strong grasp on financial analysis, real estate markets, and economic
trends.
Valuation Methods
Consider yourself before you picked this book to read. If you were asked to value a CRE asset,
what would your approach have been? There are some common ideas we hear from real estate
enthusiasts. One could say: “I will look at the price at which the asset was bought earlier.” But
what if the hotel is new, meaning it was never bought before? A logical solution, in this case, is:
“then we shall estimate how much it costs me to build the asset.” Well, if this was your line of
thought, then you were already working towards a standard method called “the cost approach.”
Some others may come up with a more practical idea: “how about looking at the price of other
similar assets sold in the market?” But what if no other asset has the same size? A possible solution
is to standardize the comparable asset prices into ratios (e.g. price per square meter). If this is
something you had thought of, then you were headed towards a method called the “sales
comparison approach”.
Some others may take a more “financial” perspective when thinking of CRE valuation.
After all, investors are focused on the cash flows that the asset will generate in the future. Given
our knowledge of cash flow analysis, we could think of an asset as a series of cash flows. From
that standpoint, the hotel is the present value of future cash flows it will generate. In real estate,
we call it “the income approach to valuation”. As the income approach to valuation is the most
widely accepted (and most popular) method of valuing CRE, let us start by exploring this.
To summarize, there are three main methods applied to value real estate assets:

Page 8 of 93
1. Sales Comparison Approach is retrospective in nature. With this approach, one assumes
that the valuation of an asset will follow a similar pattern to how market participants priced
comparable assets in the past. Some adjustments may be necessary.
2. Cost Approach to Valuation takes a more contemporary perspective. With this approach,
the value is considered close to how much it will cost to develop a similar asset at a similar
site today. Some adjustments for the aging of the property may be necessary.
3. Income Approach is prospective in nature. With this approach, one assumes that a CRE
asset could be perceived as a series of future cash flows. One derives a value estimate from
the future cash flow forecasts.
Income Approach
A CRE Asset is a Series of Cash Flows
We never said that a CRE asset is a series of “income streams”, though. For investors, CRE is
rather a series of “cash flows” for a practical reason that it is the cash flows that matter and that
cash flows and income could be substantially different in CRE settings. Financial accounting items
such as depreciation, loan amortization, reserves for replacement, etc. (we shall explore them
further in Chapter XYZ) may deviate cash flows from income. Besides, what owners will keep
aside in the name of “Capital Reserves” may not be recognized as an expense in the Income
statement. It is possible that an asset records high income in its Income (Profit & Loss) statement,
but the cash flows in the assets are low.
In financial statements (e.g., statement of cash flows), an asset may record cash flows at
different levels (asset, lender, equity investor, before tax, after-tax, etc.). In all its heterogeneity,
the CRE market broadly agrees on a specific, asset-level cash flow estimate called Net Operating
Income (NOI) which must be used for valuing CRE assets. NOI is somewhat analogous to an
asset’s free cash flow: an amount practically freed up from an asset operation (and a budget for
investing in its necessary upkeep) to be allocated to financing or taxation activities. Financing and
taxation cash flows may be different across different stakeholders in a market and may lead to
multiple cash flow estimates for the same asset. The NOI measure helps us avoid falling into this
pitfall. As NOI reflects an asset’s own attribute, we expect that the stakeholders' views on the NOI
will broadly converge within a narrow range.
In the Income Approach to Valuation, we consider two related methods, that we shall sequentially
explore:
1. Direct Capitalization
2. Discounted Cash Flows
Refer to the exhibit 5.2 below.
Suppose the current owner keeps the asset for the next N years. A holding period of N years
implicitly assumes that the asset will be sold immediately after. At the end of the Nth year, the sale
price of the asset can be determined using the Direct Capitalization approach. The net cash flow
from Sales is the “Reversion” cash flow, also known as “Cash Flow from Sales” or “Net Sales

Page 9 of 93
Proceeds” (NSP). For the first N years, the valuer will also forecast cash flows from the operations
(CFO) of the asset. In CRE, the CFO is referred to as net operating income (NOI).
For the first N years, the CFO will equal NOI. For the Nth year, the total cash flow will be
the sum of NOI and NSP. By following this process, a valuer reduces a complex asset into N yearly
cash flows. Annual cash flows mitigate the issue of seasonality in cash flows10. Finally, the valuer
calculates the GPV of these cash flows at an appropriate discount rate, which serves as the value
estimate.
Mathematically, the Value of the asset at present (𝑛 = 0) equals

𝑁
𝑁𝑂𝐼𝑛 𝑁𝑆𝑃𝑁
𝑉0 = ∑ +
(1 + Ω)𝑛 (1 + Ω)𝑁
𝑛=1
Here Ω is the discount rate derived from the market trends. As such, Ω may deviate from
an investor’s cost of raising capital. In the following mini-case (VastraMart), we discuss about
how NOI is calculated. NSP equals the sale price in year N minus net of all costs of sales (e.g.,
brokerage, consultants’ fees, lawyers’ fees, etc.):
𝑁𝑂𝐼𝑁+1
𝑁𝑆𝑃𝑁 = − 𝐶𝑜𝑠𝑡 𝑜𝑓 𝑆𝑎𝑙𝑒𝑠
𝛾𝑜𝑢𝑡
As shown above, the sale price at the end of the holding period is determined by
capitalizing the following (𝑁 + 1)𝑡ℎ year’s NOI by the going-out cap rate estimated after all
necessary adjustments 𝛾𝑜𝑢𝑡 .
VastraMart: Case Study
Consider, VastraMart a mid-tier, 3-floor shopping center in Vastrapur, Ahmedabad built in 2010.
The simple architectural design allocates a 50 sqm floor area to each shop that includes a toilet and
a kitchenette. There are eight such shops on each floor. Besides, the building has spacious common
areas such as 3 meter-wide singly-loaded corridors, 2 wide staircases, a lift lobby, and a
Security/Maintenance office. While the total rentable area is 1200 sqm, the building has 1600 sqm
of floor area in total. Thus, although the 24 potential tenants sign a 50 sqm shop each for rent, they
also enjoy the 400 sqm common area. This results in a nearly 67 sqm super area allocated to each
tenant. Tenants are charged based on the “super area.” You are analyzing the asset in 2018. By
this time, the asset has “stabilized” in that it experiences a stable growth in cash flows 11. If an
investor wanted to buy VastraMart in 2018, what would be a good price bid?
10
Some investors, especially in the private equity markets, prefer quarterly projection of cash flows. Else, in real estate
valuation, annual cash flows projections are widely acceptable.
11
By that time, the Covid -19 pandemic was not in the picture, and the market was expected to remain stable.

Page 10 of 93
Estimating the EBITDA

If VastraMart was fully occupied, how much annual rent would it collect? The rental rate is Rs.
1000/sqm/month based on the super area. This estimate called Potential Gross Income (PGI) will
be:
₹1000
𝑃𝐺𝐼 = × 1600 𝑠𝑞𝑚 × 12 𝑚𝑜𝑛𝑡ℎ𝑠 = ₹19,200,000
𝑠𝑞𝑚×𝑚𝑜𝑛𝑡ℎ𝑠
As rental rates are a result of a “price discovery process” in a market comprised of demand
and supply-side agents, they depend on market conditions12. However, in real estate markets, an
asset’s locality and quality may have a major impact on expected rental rates. A sample of 2020
rental rates in selected Indian cities is presented in the graph below. There is a notable variation
within a city:
Nevertheless, it is impractical to assume that all shopping units will always be occupied.
Some tenants may leave. Until the new tenant is found, some units may remain vacant. Besides,
searching for the tenants will come with an additional cost of brokerage, etc. Such losses in PGI
are called Vacancy loss (VAC). Suppose we anticipate a 10% loss due to tenant turnover.
𝑉𝐴𝐶 = 10% × ₹19,200,000 = ₹1,920,000
Effective Rental Income (ERI) = 𝑃𝐺𝐼 – 𝑉𝐴𝐶 = ₹17,280,000
VastraMart has a spacious parking lot on the ground floor which earns ₹10,000 on average per
weekday. Considering 6-day week and 50 open weeks in a year, this “miscellaneous” income (MI)
sums up to:
𝑀𝐼 = ₹10,000 × 6 × 50 = ₹3,000,000
PGI and MI minus net of VAC is called effective gross income (EGI).
𝐸𝐺𝐼 = 𝑃𝐺𝐼 + 𝑀𝐼 – 𝑉𝐴𝐶 = ₹20,280,000
Owners of VastraMart have several costs involved. Salaries of the security and
maintenance staff, cleaning, and utilities (e.g., electricity, water, etc.) add to the cost of common
area maintenance (CAM). Besides, they need to pay property taxes to the municipality and buy
insurance to protect against damages and liabilities. All these “operating” expenses (OPEX),
combined together are 25% of EGI.
𝑂𝑃𝐸𝑋 = 25% × ₹20,280,000 = ₹5,070,000
𝐸𝐵𝐼𝑇𝐷𝐴 = 𝐸𝐺𝐼 − 𝑂𝑃𝐸𝑋 = ₹20,280,000 − ₹5,070,000 = ₹15,210,000
Reserves for Replacement
Assets lose value due to depreciation. To keep up with the current value, one must offset the
potential depreciation through appropriate budgets. In addition to OPEX, owners also like to
maintain their property by replacing damaged or defective FF&E. One must allocate a budget for
the maintenance of existing physical assets to address the physical depreciation. These “Reserves
12
Assuming that the asset is of average quality.

Page 11 of 93
for Replacement” (CapRes hereafter) equal a certain proportion of revenues kept aside, and are
spent on an as-needed basis.
Besides, owners may occasionally make improvements to the building (e.g. façade
treatment, new security systems, new carpet, etc.) to address functional, or external obsolescence.
Functional obsolescence refers to the loss in value when an existing asset is severely outdated by
newer ones. For example, CRT TV sets were made functionally obsolete by plasma TVs. External
obsolescence relates to loss in value due to externalities. For example, a regulatory ban on alcohol
may render a bar useless. Similarly, persistent fear of the pandemic may render a spa less valuable.
Obsolescence often leads to a larger budget allocation towards capital expenditure (CAPX
hereafter). Although CAPX only occurs occasionally, it may demand a substantial budget.
Therefore, owners keep some budget aside every year in anticipation of such an expense in future
years. Several owners may keep an additional budget aside to address the anticipated CAPX. Of
course, additions (e.g. a new parking lot) or major enhancements (e.g. Converting an open space
into a swimming pool) to the assets with a view to enhance the value of an asset is also considered
CAPX.
For simplicity, we shall ignore the value-enhancement motive and assume that all sorts of
depreciation (including obsolescences) are addressed through a regular budget allocation, and club
them all as CapRes.
Usually, the CAPX budget (CapRes) depends on revenues generated. Suppose, VastraMart keeps
5% of the EGI aside in the name of CapRes.
𝐶𝑎𝑝𝑅𝑒𝑠 = 5% × ₹20,280,000 = ₹1,014,000
Estimating the NOI

These line items constitute a standard cash flow statement in CRE analysis. The NOI (=
𝐸𝐵𝐼𝑇𝐷𝐴 – 𝐶𝑎𝑝𝑅𝑒𝑠) is calculated in the following steps:
Insert Table 5.1 here
More on Capital Reserves

Capital Reserves (CapRes) is a reality: All assets will wear and tear and may warrant occasionally
major overhauls: both in the building as well as FF&E13. Besides, when a tenant terminates a lease,
the owner may have to incur leasing costs. These may include “Leasing Commissions” (LC), rent
discounts, etc. For important tenants, the owner may also be obliged to incur additional costs in
improving the facilities to meet the needs of the new tenant. Such expenses are called “Tenant
Improvement (TI).” These anticipated expenses are also added towards capital reserves. Keeping
some budget (i.e. CapRes) aside in the name of CAPX is a necessity for the smooth operation of
CRE assets. Although the capital reserves will not be counted as an expense, they will still reduce
the available cash flow. Intuitively, capital reserves must be accounted for in the cash flow measure
(NOI) for valuation purposes.
13
Furniture, fixture, and improvement

Page 12 of 93
As real as they are, the estimation of the actual CapRes is problematic. The CAPX can be
very lumpy even in stabilized assets14. Therefore, allocating an annual reserve that reflects the
expected average expenditures over the next years offers a better solution. However, CapRes
estimate is a) subjective, and b) even if it were objective, it would differ from investor to investor.
If one of the purposes of the capital reserves is to maintain the value of the real estate and FF&E,
and if the provisions for these items are adequate, then in stable market conditions, one should
expect the depreciation to be zero. Adequacy of the capital reserves can be estimated as follows:
Consider a CRE asset (including FF&E and other depreciable components) currently
valued at 𝑉0 depreciates by a factor of 𝑑 each year. After a useful life of 𝑢 years, the salvage value
𝑉
of 𝑉𝑢 is left such that 𝑉𝑢 = 𝑝. Then,
0
𝑉0 (1 − 𝑑)𝑢 = 𝑉𝑢
1
⇒ 𝑑 = 1 − 𝑝𝑢 -----------------------(1)
Suppose, an asset has a useful life of 75 years and a salvage value worth 20%. Then the
1
annual depreciation will be roughly 2% (= 1 − 20%75 ) of the asset value. Suppose, the revenues
are 15% of the asset value, then nearly 13% of the revenues must be kept aside to address the loss
in value due to depreciation15. This is clearly not the case. Most owners keep aside much less than
10% of revenues towards the CapRes provisions. Therefore, an asset will lose value due to
depreciation over time, considering normal market conditions.
Valuation Using the Direct Capitalization Method

Now, suppose that stabilized growth rate in VastraMart’s NOI is 4%. Therefore, the NOI in 2019
would be:
𝑁𝑂𝐼2019 = 𝑁𝑂𝐼2018 × (1 + 𝑔) = ₹14,196,000 × (1 + 4%) = ₹14,763,840
If the discount rate for such an investment (i.e. mid-tier shopping center in Ahmedabad) is
14%, then the value of VastraMart in 2018 would be:
𝑁𝑂𝐼2019 ₹14,763,840
𝑉2018 = = = ₹147,638,400 -------------------------------(2)
𝑟−𝑔 14%−4%
The above derivation of value (i.e. approx. ₹15 crores) is simply an application of the
constantly growing perpetuity mathematics that we discovered in Chapter 4. Congratulations! You
just valued the first CRE asset using a scientific method called Direct Capitalization. Yet, before
you get too excited about your CRE valuation skills, here is a word of caution: The Direct
Capitalization method of valuation is based on a strong assumption that the cash flows have
stabilized. In reality, cash flows rarely stabilize perfectly. Due to market cycles, and some
14
i.e. some years may have a very high CAPEX than some others.
15 𝐶𝐴𝑃𝑅𝑒𝑠 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 𝐶𝐴𝑃𝑅𝑒𝑠 2%
If = 2%, = 14%, then = = 13.33%,
𝑉 𝑉 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 14%

Page 13 of 93
randomness, the future cash flows may fluctuate and deviate from their expected (“constantly
growing”) values, even if the business has matured. Thus, this method is our best-educated guess
on valuation. Yet, this is a critical building block for CRE valuation.
At this stage, one pertinent question to ask is whether ₹15 crores is the market value (MV)
of VastraMart, or its investment value (IV) specific to an investor. The answer lies in the
assumptions about the discount rate (𝑟) and growth rate (𝑔). These numbers may be different for
different investors. If 𝑟 is derived from an investor’s cost of capital (WACC), it will depend on
their preferred capital structure, access to capital, and cost of capital. All these numbers will vary
across investors. Similarly, 𝑔 will depend on how aggressive an investor’s business plan is.
Clearly, VastraMart may imply different IVs for different investors. If our intent is to estimate the
MV, we need to be agnostic towards individual preferences and think in terms of the market’s
perception of these numbers.
Capitalization Rate
From our knowledge of the components of return, we can say that 𝑟 is the expected total return, or
expected IRR of VastraMart; and 𝑔 is the expected income return or “yield.” If the denominator
in Eq(1) is broadly stable, then the expected growth in cash flows (𝑔) is also the expected16 growth
in value, commonly known as the “appreciation return” or “gains.”
∴ 𝐸[𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑡𝑢𝑟𝑛 = 𝐼𝑅𝑅] = 𝐸[𝐼𝑛𝑐𝑜𝑚𝑒 𝑅𝑒𝑡𝑢𝑟𝑛] + 𝐸[𝐴𝑝𝑝𝑟𝑒𝑐𝑖𝑎𝑡𝑖𝑜𝑛 𝑅𝑒𝑡𝑢𝑟𝑛]
➔ 𝑟 = 𝑦 + 𝑔
Therefore, the denominator in Eq(1) i.e. 𝑟 − 𝑔 reflects the yield expectation, commonly known
as Capitalization Rate (or “Cap Rate” in short) in real estate17. Our equation for valuation using
the Direct Capitalization method reduces to
𝑁𝑂𝐼(𝑡+1)
𝑉𝑡 = , where subscript (𝑡 + 1) is the index year and 𝑦 is the cap rate. ---------------------(3)
𝑦𝑡
When estimating MV, the cap rate must be dictated by the market. In active real estate
markets, consulting firms frequently survey market participants on their perception of the cap rate.
For example, in the US, firms such as RERC, IRR, and organizations such as NCREIF collect the
market-reported cap rate (across asset classes and locations) each quarter. In India, most real estate
consulting firms tend to keep this information proprietary. This tendency has added to the opacity
of real estate markets, and international investors tend to be skeptical of valuations. It is
unfortunate that, unlike sophisticated real estate markets, only a minuscule segment of real estate
firms in India appreciate the importance of sharing data for academic research. However, some
rating agencies and research firms in India have started publishing cap rate surveys at annual levels.
Note that, while for valuers, a cap rate is an input for a valuation model, the cap rate does
not “cause” the valuation. When survey respondents report their perceived cap rates (which the
16
E[X] is a notation used to express the expected value of x. If x follows a probability distribution, E[x] equals the
mean of the distribution
17
Over time, Capitalization (cap) rate has gained a specific connotation in real estate circles. This is not to be confused
with the “capitalization rate” used in general/corporate finance jargon, where it simply refers to the opportunity cost
of capital (OCC) or discount rate. Yes, sometimes jargons are more confusing!

Page 14 of 93
consulting firms abstract and publish as an average), the cap rate is an outcome of perceived asset
value, an outcome of their value judgments. Nevertheless, for the valuation community, the market
cap rate is a precursor to value estimation. The application of cap rate as an input to valuation
makes a rational sense: a valuer’s task is to estimate the most likely price at which an asset will
sell in the market. By applying the market cap rate as an input to valuation, this purpose is well
solved.
“… Study the science of art; study the art of science. Learn how to see. Realize that everything connects
to everything else.” — Leonardo da Vinci
It is difficult to definitively tell how the cap rate survey respondents provide their cap rate
estimates. Perhaps, some respondents endeavor to estimate the cap rate by applying the terminal
rate formula: 𝑦 = 𝑟 − 𝑔. This alternate method demands significant market research, as it
requires estimating the WACC of a typical hotel investment firm as well as an educated analysis
of projected growth in the future cash flows. This task is tedious and fraught with errors. Plausibly,
such respondents will use a great deal of their “gut feelings” to estimate the cap rate.
Alternatively, the respondents privy to recent transactions may use EQ(3) to report the cap
𝑁𝑂𝐼(𝑡+1) 𝑁𝑂𝐼(𝑡+1)
rate: 𝑉 𝑡 = 𝑦 → 𝑦𝑡 = 𝑉 . One can assume that such respondents have a good sense of
𝑡 𝑡
the projected NOI. They would still need the value of their assets (V) to estimate the cap rate. In
relatively fewer cases, when an asset was transacted, the transaction price could be used as the
𝑁𝑂𝐼(𝑡+1)
value; and the reported cap rate will equal 𝑦𝑡 = 𝑃 , where P is the transaction price of the
𝑡
asset. Such a cap rate is a “transaction bases” cap rate. Unfortunately, transactions are infrequent,
but the appetite for cap rate estimates over time is everlasting. Therefore, many respondents must,
again, use their “gut feelings” on asset values.
Then, there is a third group of respondents who are not privy to any transaction or valuation
information. Their perception of cap rate is purely driven by macroeconomic environments. The
association between such an environment and asset prices, cash flows, growth, and cost of capital
is imperfect. For such respondents, the role of “gut feelings” is even stronger in cap rate estimates.
The result is that the market cap rate benchmarks available from market reports are broadly results
of a heuristic process, a process fraught with mental short-cuts and behavioral biases. So long as
the surveys converge towards a consensus on cap rates, they are important to valuation. Whether
a market is rational or not, the opinion of market players matters immensely in valuation. After all,
market value is the estimate of the most likely price these market players would be willing to pay
for an asset. No wonder, Menger’s “Subjective Theory of Value” is still in vogue. If the market value
is the true value, all our scientific endeavors go into guessing the thought process of market players
which is prone to behavioral biases and is no pure science. They must render the expected price
for an asset to be its true value. They apply their scientific tools to reach at a number (the value
estimate) which is a result of perceptions. Valuers must make their peace with this reality:
valuation is an art supported by scientific tools.
Nature of Capitalization Rate

As an input to valuation models, the cap rate serves as an expected income return. As a return
measure, it must be directly proportional to the perceived risk in an asset. Have another look at

Page 15 of 93
EQ(3). Note that the cash flow (NOI) in the numerator evolves in the space market, the market for
operating an asset whereas Value (V) is more closely tied to the asset market, the market for
investors in CRE. This market primarily comprises institutional investors, such as funds and
portfolio managers. These investors are more interested in raising and deploying capital to generate
returns. Thus, their choice of an operator of CRE is expected to be optimal that decides on an
expected NOI. Considering that investment and valuation decisions depend on a given value of
NOI, cap rate must be inversely proportional to value:
1
y∝𝑉
This is an important concept. Intuitively, a high cap rate implies higher perceived risk, and
hence a higher expectation of return. For a given stream of future cash flows, an investor must
maximize her return by minimizing the investment (i.e. the price of the asset = value) in the present.
Market cap rates are directly proportional to the perceived risk in an asset class. Historical cap
rates for various asset classes in selected cities of India are plotted below:
The plot suggests that cap rates may vary across locations and asset classes. For example,
retail assets in Mumbai have had substantially low cap rates since 2012 than all other market
segments whereas retail assets in Chennai have had the highest cap rates. Cap rates may also vary
over time within a market segment. For example, retail cap rates in Gurgaon have witnessed a
significant ebb and flow during the six years plotted in the graph. Note that due to data scarcity,
these cap rate curves may have been based on fewer transactions and, as a result, could be noisy.
With a larger set of observations, we should expect these curves to be smoother. Nevertheless, cap
rates in CRE assets in India tend to be on a higher side when compared with international markets.
For example, a CBRE Report (2020) suggested that average cap rates for core office properties in
some Asian markets (Hong Kong, Taipei, Tokyo, Singapore, etc.) were as low as 2-3%. Beijing,
Shanghai, Shenzhen stood at 3-4%, Guangzhou, Sydney, Melbourne at 4-5%. Cap rates in Indian
markets (Gurgaon, Mumbai, Bengaluru) were relatively high, nearing 8%.
From more established markets, we can observe the following broad trends in Cap Rates18:
1. Perceived Risk: Increase in perceived risk is associated with an increased cap rate and vice
versa.
2. Locality spreads: Unless a location experiences a drastic makeover (or distress), cap rate
spreads across localities tend to be stable over time. Localities that are growing fast, or
those which have matured economy, will have lower cap rates. Also, localities with fewer
constraints to new supply (e.g., easier project approval) tend to have higher cap rates.
3. Booming markets: When an economy is booming, the demand for assets increases, often
more than the promised cash flows. As a result, there is an upward pressure on asset prices
leading the cap rates to fall.
4. Bust markets: When the markets start to crash, fewer transactions can be observed. As a
result, in lack of transactions, asset values tend to be relatively sticky in the first few
18
See https://hotelnewsnow.com/Articles/30380/7-tips-to-adjust-a-hotels-benchmark-cap-rate

Page 16 of 93
quarters, although the cash flows fall substantially. As a result, the cap rates may appear to
fall initially, which is counterintuitive. Gradually, the supply of assets listed for sale swells,
and opportunistic investors start to buy assets. The fall in prices is more drastic than the
fall in cash flows. As a result, a few quarters in a crisis time after it starts, the cap rates tend
to creep upwards.
5. Normal markets: In persistently long normal market conditions, we should expect the cap
rates to remain stable within a narrow range. However, depending on the location,- the
demand for certain assets may continue to increase (or decrease), and a secular trend in the
cap rates may still be observed. The deviation from the trend usually stays within 25 bps.
In the short run, shocks (e.g., sudden increase or decrease) tend to continue over
consecutive quarters. However, cap rates will self-discipline themselves with changes in
the opposite direction within a year to come closer to the trend.
6. Inter asset class/subclass spreads: Unless something drastic happens within a market, the
perceived risk across asset classes remains stable. As a result, the spread across different
asset classes tends to be stable over time. For example, during recent decades in the US,
the contemporaneous cap rate spread across hotels and rental apartments varied between
250 to 300 basis points. Across different tiers of an asset class (e.g., high-end versus low-
end offices, or luxury versus mid-scale residences), the spread tends to be small, usually
within 100-150 basis points.
Estimation of Cap Rate

We cannot stress enough on the requirement that for a market value estimate of an asset, the
capitalization rate must be dictated by the market. Market reports are the best sources of
capitalization rate benchmarks. In some markets, this information may be made available for free
by some consulting firms. For more precise benchmarks, one may have to subscribe to paid reports.
Yet, even in more sophisticated markets (e.g., US or UK) few markets and fewer CRE segments
within them are covered in these reports.
A valuer must deal with the humongous challenge of limited information and parse through
subjective signals to translate them into quantifiable metrics. This is time we shredded our toga of
applied scientists and move on to the art of valuation, to a territory of thumb rules which are
somewhat scientific, but not purely.
When a Cap Rate Benchmark is Available

For a cap rate estimate, it is best to start from a market benchmark. In the most likely scenario,
when the benchmark is not the most suited one for the asset to be valued, the benchmark cap rate
must be adjusted using the nature of cap rates. These adjustments are based on broad thumb rules,
but often work well in real-life situations. We suggest following two simple rules in cap rate
adjustments from a benchmark:
1. With an increase in perceived risk, adjust the cap rate up. Similarly, with a decrease in
perceived risk, adjust the cap rate down. The risk differential will be priced in terms of a
few basis points in the cap rate adjustment.
2. Keep the adjustment at each level within a modest range. A ±25 to 50 bps adjustment is
often substantial to adjust for risk differential. In rare cases, these adjustments could go as
high as 100-150 bps. To be less “pretentious” in the adjustments, we suggest making

Page 17 of 93
adjustments in multiples of 25 bps. After all adjustments, it is good to further round the
final cap rate judgement as a multiple of 25 bps19.
First, try to find the closest cap rate benchmark available to you. Then make adjustments, in
steps, based on the nature of the capitalization rate discussed above. Here are the criteria for
picking a benchmark:
1. Location: The benchmark should be from the same location. In some cases, when based
on past observations you are convinced about the similarity of cap rates and their co-
movements, a benchmark from another location may be suitable too. For example, office
capitalization rates in Hyderabad and Gurgaon seem to be within very close range of each
other, despite being geographically apart. However, when picking the cap rate benchmark
from a different location, you must convince the audience with data on why the cap rates
in these two markets should be similar at a given time.
Example 1: You have reasons to believe that Pune office markets follow a similar
trend as Mumbai but are riskier than Mumbai. However, Pune office markets are,
hypothetically, less risky than Bangalore office markets. Based on this set of
information, you know that the Pune office cap rate in 2015 must have been within
the range of Mumbai (7%) and Bangalore (9%). A logical estimate would be 8%
for Pune offices. However, one could argue that the Pune market is more like
Mumbai than Bangalore. As such, a better estimate of the Pune office cap rate
would be 7.5%.
Example 2: In 2015, Gurgaon retail assets were trading at a cap rate of 8.5%.
Similar retail markets in New Delhi would trade at a cap rate of 8% or 8.25%
considering a relatively lower risk. One could use similar adjustments within
different submarkets of Gurgaon.
2. Asset Class: Due to limited data, we do not quite observe a similar pattern in cap rate
across Indian cities, however, in stable and matured CRE markets, it may be fair to assume
that the cap rate spreads across two asset classes (e.g., hotels versus offices) would broadly
stay the same over normal market conditions.
Example: Some luxury hotels and offices in South Delhi were sold at cap rates of
7% and 8.5% respectively. In 2019, similar quality offices were selling at a cap rate
of 7.5%. For 2019, one could apply the same spread of 150 bps to conclude that
luxury hotels would sell at a cap rate of 6%.
3. Time: Risk perception within an asset class and location may change over time. During
crisis periods, the time variation may call for drastic adjustments. Else, given the sticky
nature of cap rates, these adjustments could be carried out in a similar fashion.
Example: During 2018 Q4, a specialty restaurant in Udaipur sold at a cap rate of
10%. In 2019 Q4 due to some concerns about an upcoming pandemic, the market
was slightly riskier. The cap rate for a similar restaurant in Udaipur could be
adjusted up to 10.25% or 10.5% in 2019 Q4.
19
In EXCEL you could use MROUND() function to achieve this. But manual adjustments work just fine.

Page 18 of 93
4. Asset Specificity: Cap rate benchmarks represent a typical asset within its class and
location, whereas the subject asset may be significantly different.
Example: A resort close to an urban castle in Jodhpur may be less risky than a
resort on the outskirts of the city. However, unless the differences are very
convincingly established, the adjustments made should be small, e.g. 25 bps.
Note that the thumb rules presented above are very broad in nature and an analyst must use
relevant information and common sense to break away from these thumb rules. If no more accurate
information is available, these thumb rules offer a scientific way of adjusting the cap rate estimate.
In an upcoming mini case “Jane Decides on Cap Rate”, we shall reconcile our understanding of
cap rate adjustments into a real-world context.
When a Cap Rate Benchmark is Not Available

Our discussion above assumed that we shall have a cap rate benchmark available to adjust from.
However, in most parts of the world (including India), the availability of the cap rate benchmark
is a luxury. In such a situation, talking to experienced CRE valuers in a market may offer a good
alternative for cap rate benchmarks. In less liquid markets, expert opinion may influence investor
perception of valuation. But in a less active CRE market, such experts may be a luxury too. In
those cases, we advise alternative methods of cap rate estimation as listed below, in their order of
priority, as follows:
First Principles Approach

An analyst could collect relevant market information to pose as the typical investor in the market.
Appropriate data could be used to estimate the WACC of a typical investor in an asset class.
Similarly, the growth rate in the cash flows could be estimated independently. The difference of
these two (i.e., WACC -g) could serve as a proxy for the capitalization rate.
Example: Jawed surveyed owners of retail shops in suburban Vizag. On average, their
investments were financed with 60% bank loans that charge a 10% interest rate. The equity
investors expect a 15% return from such investments and expect a 4% growth in annual
rent. The cap rate (𝛾) can be estimated as follows:
𝛾 = 𝑊𝐴𝐶𝐶 − 𝑔 = [𝑤𝑒 × 𝐾𝑒 + 𝑤𝑑 × 𝐾𝑑 ] − 𝑔
= 40% × 15% + 60% × 10% − 4% = 8%
Band-of-Investment Approach
A relatively outdated “Band of Investment” (BoI) approach could serve to be useful. This is
somewhat similar to the first principles approach, as it appears to be positing the cap rate as a
special type of WACC that focuses on yield expectations only (and ignores the growth
component). As cap rate (𝛾) is the income component of return, this method may be applied as
follows:
𝛾 = 𝑒𝑞𝑢𝑖𝑡𝑦 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 + 𝑓𝑖𝑛𝑎𝑛𝑐𝑖𝑛𝑔 𝑐𝑜𝑚𝑝𝑜𝑛𝑒𝑛𝑡 = 𝑤𝑒 × 𝑦𝑒 + 𝑤𝑑 × 𝑦𝑑 ,

Page 19 of 93
where 𝑤 refers to the weight of capital, 𝑦 refers to expected yield (income return), and 𝑑 and
𝑒 index debt and equity capital, respectively. The yield on equity is derived from the expected
dividend yield by equity investors in similar investments (and excludes the appreciation return).
The traditional formula uses mortgage constant as 𝑦𝑑 , however, a more accurate measure for 𝑦𝑑 is
the effective cost of debt20.
Example: In the above example, assume that investors expect a 6% equity yield. The cap
rate (𝛾) can be estimated as follows:
𝛾 = 𝑤𝑒 × 𝑦𝑒 + 𝑤𝑑 × 𝑦𝑑
= 40% × 6% + 60% × 10% = 8.4%
Note that in a traditional BoI approach 𝑦𝑑 will equal the mortgage constant and overstate
the cap rate21.
Risk-Reward Approach
A theoretically sound but (often) practically messy method of estimating cap rates is to think in
terms of risk and reward. If the cap rate is related to return expectation, it should be directly
proportional to risk. Therefore, the cap rate should be equal to the risk-free rate plus the risk
premium.
Example:
Theoretically, indeed, it is a sound argument. In the above example, an analyst identified seven
different sources of risk, priced them in terms of risk premia, and summed them to the risk-free
rate. As elegant as this method appears, it is often far from being so, scientifically. There are two
challenges that lie in (1) identifying and (2) pricing the sources of risk. What if some of the
identified risks do not matter? What if some important sources of risk were omitted from the list?
How can we be sure about how much risk premium is warranted against each source of risk?
As of now, all we could tell about cap rate (𝛾) in relation to the risk-free (RF) rate is as follows:
𝛾 = 𝑅𝐹 + 𝑅𝑃,
where 𝑅𝑃 is a theoretical measure of risk premium which can be observed as an aggregate of
all risk premia. However, this equation must not mislead one into thinking that the risk-free rate
and cap rate co-move with each other. This would imply that 𝑅𝑃 is, at least broadly, a constant. It
is not. During a downturn when the perceived market risks are high, central banks tend to cut down
the risk-free rate to force the interest rates down. In several cases, a significant cut in 𝑅𝐹 does not
effectively influence market players’ perception about CRE. As a result, the cap rate may continue
20 𝑎𝑛𝑛𝑢𝑎𝑙 𝑚𝑜𝑟𝑡𝑔𝑎𝑔𝑒 𝑟𝑒𝑝𝑎𝑦𝑚𝑒𝑛𝑡 𝑎𝑚𝑜𝑢𝑛𝑡

Mortgage Constant = . As you will see in Chapter 6 that, the numerator in the
𝑙𝑜𝑎𝑛 𝑎𝑚𝑜𝑢𝑛𝑡
equation could also include loan principal. As a result, the mortgage constant may overstate the 𝑦𝑑 .
21
Considering a 20-year, monthly amortizing loan, the mortgage constant will be 11.6%, that will lead to a cap rate
estimate of 9.36%

Page 20 of 93
to creep up rendering risk premia to be highly volatile and unpredictable. If one cannot predict the
𝑅𝑃, the risk-reward approach to cap rate estimation ends up being a theoretical concept with
limited practical use. Yes, some extremely wise and experienced analysts may still be able to
estimate 𝑅𝑃 more accurately than others. The following graph shows how the risk-premium
(Spread of cap rate from interest rate) in cap rate across selected office markets in India evolved
between 2010-2016.
Similar phenomena can be observed in the risk premium across the globe as shown below:
Cap Rate Adjustment for Capital Reserves

The property-level cash flow is an essential measure for determining the cap rate of an asset, given
its value. If the cap rates reported from the market surveys are to be applied, one needs to ensure
that all the parties have the same vision of the cash flow. NOI’s primary appeal as the cash flow
measure used in CRE valuation lies in its characterization by the underlying asset. NOI is an asset-
level measure of cash flows, that does not account for potential loan payments or tax payments
that may differ across investors. More importantly, NOI is based on the performance and
characteristics of an asset, not on capital market decisions of an owner, such as loans repayment
or dividend payout, etc. In short, NOI is a real sector measure rather than a financial sector
measure.
Capital reserves are a real sector consideration and must be included in the cash flows of
an asset when valuing it. The problem arises when different owners of the same asset may have a
different capital reserve budget. Unfortunately, this is often the case. For these reasons, the NOI
measure fails to limit room for investors to disagree on the numbers. That would mean that the
same asset may lead to different value estimates by different analysts, depending on their estimate
for CapRes.
Using an asset’s EBITDA could help resolve the issue as it ignores CapRes. Indeed, many
analysts posit EBITDA as the NOI. For our discussion, let us call it “𝑁𝑂𝐼 𝑏𝑒𝑓𝑜𝑟𝑒 − 𝐶𝑎𝑝𝑅𝑒𝑠” (=
𝑁𝑂𝐼𝑏 ). Several other analysts would rather account for CapRes and calculate the “𝑁𝑂𝐼 𝑎𝑓𝑡𝑒𝑟 −
𝐶𝑎𝑝𝑅𝑒𝑠” (= 𝑁𝑂𝐼𝑎 ). The difference between 𝑁𝑂𝐼𝑏 𝑎𝑛𝑑 𝑁𝑂𝐼𝑎 goes beyond the nomenclature. The
CapRes often ranges between 10%-20% of EBITDA in typical settings. Thus, the 𝑁𝑂𝐼𝑏 𝑎𝑛𝑑 𝑁𝑂𝐼𝑎
will be 10%-20% apart from each other. The challenge is that for a given cap rate estimate, the
resulting valuation, too, will be sensitive to how a valuer defines the NOI.
Which one is a more appropriate measure of the NOI in reporting the cap rates? This
confusion also creeps into various cap rate surveys. Industry reports estimate that nearly 40% to
80% of the survey respondents report the cap rates based on 𝑏𝑒𝑓𝑜𝑟𝑒 − 𝐶𝑎𝑝𝑅𝑒𝑠 NOI. Institutional
investors seem more inclined towards this measure.
For a hotel with a certain value, the cap rates based on before (or after) CapRes deduction
will be different to adjust for the different cash flows corresponding to the CapRes consideration.

Page 21 of 93
Cap rates based on 𝑁𝑂𝐼𝑏 will be higher than the cap rates based on 𝑁𝑂𝐼𝑎 (𝑖. 𝑒. 𝛾𝑏 > 𝛾𝑎 ). Although
not without errors, it is more accurate to assume that the cap rates reported in the survey are
dominantly the 𝑏𝑒𝑓𝑜𝑟𝑒 − 𝐶𝑎𝑝𝑅𝑒𝑠 cap rates ( 𝛾𝑏 ).
Of course, one practical solution for valuers would be to ignore the CapRes altogether and
directly use the asset EBITDA (𝑁𝑂𝐼𝑏 ) for valuation purposes. This goes against the principle that
CRE valuation is derived from its cash flows. Also, for a more realistic cash flow projection,
especially for internal or investment valuation of a hotel, including the CapRes in the cash flow
estimation is required. As a result, one estimates the 𝑎𝑓𝑡𝑒𝑟 − 𝐶𝑎𝑝𝑅𝑒𝑠 cap rate (𝛾𝑎 ) based on the
𝑏𝑒𝑓𝑜𝑟𝑒 − 𝐶𝑎𝑝𝑅𝑒𝑠 cap rate estimate. These adjustments are particularly critical in lower-quality
CRE segments with high cap rates and high CAPX needs.
𝑁𝑂𝐼𝑏 − 𝐶𝑎𝑝𝑅𝑒𝑠 = 𝑁𝑂𝐼𝑎
𝐶𝐴𝑃𝑅𝑒𝑠
Suppose, = 𝑥 → 𝑁𝑂𝐼𝑏 (1 − 𝑥) = 𝑁𝑂𝐼𝑎
𝑁𝑂𝐼𝑏
𝑁𝑂𝐼𝑏 𝑁𝑂𝐼𝑎
Now, 𝑉 = =
𝛾𝑏 𝛾𝑎
𝐶𝑎𝑝𝑅𝑒𝑠
→ 𝛾𝑎 = 𝛾𝑏 (1 − 𝑥); 𝑥=
𝐸𝐵𝐼𝑇𝐷𝐴
In the VastraMart case discussed earlier, 𝑥 = 6.7%. Note that although CapRes is
determined from EGI, here we are more interested in seeing it as a fraction of EBITDA. Now,
suppose that the market survey suggests a cap rate of 10.7%.
𝛾𝑏 = 10.7% → 𝛾𝑎 = 10.7%(1 − 6.67%) = 10%
𝑁𝑂𝐼𝑏 𝑁𝑂𝐼𝑎 ₹16,506,000 ₹15,405,600
𝑉= 𝑜𝑟 i.e. 𝑜𝑟 = ₹154,056,000
𝛾𝑏 𝛾𝑎 10.7% 10%
The calculations above illustrate that EBITDA should be capitalized by 𝛾𝑏 or 𝑁𝑂𝐼𝑎 should
be capitalized by 𝛾𝑎 for an accurate valuation. If the CapRes is 15% of the EBITDA and the market
survey suggests a cap rate (𝛾𝑏 ) of 8%, the adjustment will be roughly 120 basis points. In other
words, a valuer who incorporates the CapRes in the NOI estimate must adjust the cap rate (𝛾𝑎 )
down to 6.8%. Failing to make such an adjustment implies a 20% error in value judgment. To put
things in perspective, the value of a ₹5 crore restaurant hotel will be underestimated approximately
at ₹4 crores.
As discussed earlier, capital reserves may serve two distinct purposes: (1) reserves for
FF&E replacement, and (2) capital improvements to the real estate assets. The replacement or
maintenance of FF&E is a more frequent and almost an ongoing process. Capital improvements to
real estate assets (e.g., new roofing, new HVAC system, etc.) may call for significantly higher
budgets and would only be considered once in several years. Often, a property operator must
negotiate for these two types of budgets independent of each other. Nevertheless, provisioning for
both types of budgets over time is a prudent CRE practice.
Going-In versus Going-Out Cap Rate

If cash flows in an asset have stabilized, the direct capitalization method offers a powerful tool to
value the asset. The cap rate used for such a valuation that we have been discussing so far is derived

Page 22 of 93
from the information available today. We shall say that all the adjustments so far were for the
“Going-In” cap rate (𝛾𝑖𝑛 ). Differently stated, 𝛾𝑖𝑛 is the cap rate estimated for the present time.
Sometimes, the stabilization may take several years. In such cases, we can only apply the direct
capitalization method for valuation after several years “out” in the future when we are confident
about the stabilization of the cash flows. The cap rate applied to stabilized cash flows in the future
is called the “Going-Out” cap rate (𝛾𝑜𝑢𝑡 ). Will 𝛾𝑜𝑢𝑡 be the same as 𝛾𝑖𝑛 ? Real estate experts say,
No.
In stock valuation, a constantly growing perpetuity of dividends can be capitalized by the
same rate at any time in the future to estimate the “terminal value.” The terminal rate is estimated
as [𝑟 − 𝑔] where 𝑟 is the opportunity cost of capital (“cost of equity”) and 𝑔 is the expected growth
rate in dividends. Stock price, dividend, and dividend growth rate in stock markets are analogous
respectively to asset price, NOI, and growth rate in NOI in the CRE market.
One may argue that as 𝑟 and 𝑔 are assumed to be fixed at the outset, a stock’s terminal rate
(or a CRE’s 𝛾𝑜𝑢𝑡 ) should also be fixed at [𝑟 − 𝑔]. This argument holds for stocks, but not for CRE
assets. Real estate experts argue that, unlike stocks, a CRE asset will depreciate over time. As a
result, assuming all else to be the same (and market conditions to be stable), a CRE asset’s 𝛾𝑜𝑢𝑡
will tend to grow larger than 𝛾𝑖𝑛 .
Impact of Asset Age
This implies that although two high-grade office towers in the same locality of a city may promise
to generate similar cash flows, the older one will be valued less if the age differential is clearly
observed. In some instances, it has been observed that the asset cash flows do not follow a linear
association with age. The cash flows first increase with the age, as the asset “seasons” and gains
recognition over time; but starts to fall later on. Barring the first few years of seasoning, the
decrease in cash flows, as well as the increase in the cap rate, will render older hotels to have lower
values, all else being equal. Thus, the impact of age on valuation is two folds: One via the cash
flows and the other via the cap rate.
Industry Practice
The discussion above suggests that in normal market conditions, 𝛾𝑜𝑢𝑡 > 𝛾𝑖𝑛 owing to an increase
in the perceived risk caused by ageing of the asset. Industry practitioners often assign a spread (∆)
to 𝛾𝑖𝑛 to estimate the 𝛾𝑜𝑢𝑡 such that 𝛾𝑖𝑛 + ∆= 𝛾𝑜𝑢𝑡 as shown in an example below. Beyond
functional obsolescence, all additional sources of additional risk in the future are priced in the
delta.
Some Words of Caution
If the market conditions are stable, adding some delta (spread) to 𝛾𝑖𝑛 is, at best, a rough
approximation of where the capitalization rate could be in the future given the obsolescence in the
asset. As such, the farther out in the future, the larger the spread should be. However, as we shall
see in an upcoming discussion, faced with uncertainty related to how far out in the future should
the 𝛾𝑜𝑢𝑡 be estimated for, valuers follow simple thumb rules for delta (e.g., ∆ = 25 to 100 bps).
This may lead to valuation errors and we must correct for them.
By now, we realize the following about the cap rate:

Page 23 of 93
1. 𝛾𝑖𝑛 evolves over time due to market conditions.

2. ∆ may increase over time for a given asset in stable market conditions due to an asset’s
growing obsolescence.
By fixing the spread (∆) between 𝛾𝑜𝑢𝑡 and 𝛾𝑖𝑛 at the outset, one ignores both the statements
presented above. Even in a stable market condition (wherein the market cap rates stay flat), ∆ will
be material only over longer horizons. An asset will not be faced with obsolescence in the shorter
run. Besides, valuers must consider the trends and cyclicality of market cap rate as well as the
length of time in the future. Therefore, a more accurate derivation of the going-out cap rate must
consider its own time dependence as well as the time dependence of the going-in cap rate as
follows:
𝛾𝑜𝑢𝑡,𝑡 = 𝛾𝑖𝑛,𝑡 + ∆𝑎𝑔𝑒𝑖𝑛𝑔,𝑡 .
Here, 𝑡 indexes the time in the future for which 𝛾𝑜𝑢𝑡 is being estimated. As such, the spread
∆ can be broken into two components: (1) the expected change in the market cap rate (𝛾𝑖𝑛 ) over
time due to trends and cyclicality, and (2) the time-dependent ageing effect causing obsolescence:
∆ = ∆𝛾 + ∆𝑎𝑔𝑒𝑖𝑛𝑔
If we apply the direct capitalization method for some time in the future, we need to forecast
what the market cap rate would be at that time (say, year-t). This will be the going-in cap rate in
year-t future. 𝛾𝑖𝑛𝑓𝑢𝑡𝑢𝑟𝑒 = 𝛾𝑖𝑛 𝑝𝑟𝑒𝑠𝑒𝑛𝑡 + ∆𝛾 . In stable market conditions, we should expect ∆𝛾 to be
negligible. However, in several locations, relatively stable market conditions may still witness a
secular trend in cap rates. Further, by year-t, the asset being analyzed will have aged. Therefore,
the going-out cap rate should ideally be estimated as:
𝛾𝑜𝑢𝑡 = 𝛾𝑖𝑛 + ∆𝛾 + ∆𝑎𝑔𝑒𝑖𝑛𝑔,𝑡 .
Example 1: For the last several years, office markets in Hyderabad have seen a flat cap
rate of 7%. This is expected to remain the same in the upcoming several years. Roughly
every ten years, the construction materials and FF&E go through a major upgrade,
resulting in 50bps higher cap rates in older properties. What will be the going-out cap rate
around ten years later?
In this example, ∆𝛾 = 0; 𝑎𝑛𝑑 ∆𝑎𝑔𝑒𝑖𝑛𝑔 = 50 𝑏𝑝𝑠
𝛾𝑜𝑢𝑡 = 7% + 0.5% = 7.5%
Example 2: For the last several years, office markets in Noida have seen a secular trend of
-10bps each year in cap rate which currently stands at 9%. This is expected to remain the
same in the upcoming decade. Roughly every ten years, the construction materials and
FF&E go through a major upgrade resulting in 50bps higher cap rates in older properties.
What will be the going-out cap rate around five years later?
50 𝑏𝑝𝑠
In this example, ∆𝛾 = −10 𝑏𝑝𝑠 ∗ 5 = −50 𝑏𝑝𝑠; 𝑎𝑛𝑑 ∆𝑎𝑔𝑒𝑖𝑛𝑔 = = 25 𝑏𝑝𝑠
2
𝛾𝑜𝑢𝑡 = 9% − 0.5% + 0.25% = 8.75%

Example 3: This is the fourth quarter of 2020. Industrial cap rates in the NCR area have
remained stable at 6% until 2019. In 2020, due to the Covid-19 concerns, the cap rate shot

Page 24 of 93
up to 9%. As the world finds a solution for the crisis, the risk perception will gradually fall,
and the cap rate is expected to be back to 6% in 5 years. Roughly every ten years, the
construction materials and FF&E go through a major upgrade resulting in 30bps higher cap
rates in older properties. What will be the going-out cap rate around five years later?
30 𝑏𝑝𝑠
In this example, ∆𝛾 = −300 𝑏𝑝𝑠; 𝑎𝑛𝑑 ∆𝑎𝑔𝑒𝑖𝑛𝑔 = = 15𝑏𝑝𝑠
2
𝛾𝑜𝑢𝑡 = 9% − 3% + 0.15% = 6.15%

Much discussion has gone on capitalization rate in this chapter so far. Let us reconcile our
understanding so far. The following flowchart presents the process of determining cap rate for
CRE valuation:
Case Study: “Jane Decides on Cap Rate”

Jane is a CRE appraiser. She has been commissioned by McDavid’s to appraise one of the
restaurants in Denver they are interested to buy. She observes that another restaurant owned by
BurgerBoss in the same locality just sold for $1.5 million. Jack, an analyst from Analytica Real
Estate Services, who represented both the buyer and the seller, said it was a comparably lower
cap rate he has seen for a BurgerBoss property in the region; perhaps because of higher fast-
food sales in the locality. The buyer paid the sales price for an expected income stream of
$82,500. Compared to the BurgerBoss property, the subject property has a marginally higher
risk.
In an active market, cap rates extracted from recently completed transactions (which are
commonly known as market cap rates) can provide investors and appraisers a useful guide for
determining the appropriateness of the cap rate being used to value a subject property. One
should note that market cap rates represent an average of numerous property transactions that
can vary widely according to their risk profiles. Thus, it is important to make upward or
downward adjustments to market cap rates when estimating the value of a given investment
property.
The current prevailing market cap rate for all real estate asset types at the national level is
7.75%. The Denver market capitalization rates are usually higher than the national average and
their prevailing rate is 9.25%. The national market cap rates for restaurant companies in this
period is 6.5%. Your subject property has substantially lower risk compared to market average.
• What could be the band of the going-in cap rate for the subject property?
• What could be the going-in cap rate that Jane uses for her valuation?
The investor is planning to hold the property for five years. The NOI is expected to be stable
across the investment horizon. The hotel transaction market is expected to be affected by the
huge sway from business customers to leisure customers across all segments and in all regional
markets.
• What could be the expected going-out cap rate?

Page 25 of 93
Solution
Note that this hypothetical case study is based on real data. Like most case studies, this case
study does not have a right or wrong answer and involves subjective judgment. As such, the
prescribed solution is not very scientific. We attempt to illustrate in a real-world setting:
1. How to work around the issue of limited data
2. How to quantify subjective information
The objective is to estimate the cap rate for our subject property (McDavid’s), whereas
we have some information available on our comparable property BurgerBoss as well as we
have access to some broad market level information.
Lower Limit
The comparable restaurant, BurgerBoss recently sold for $1.5 million. Its projected NOI for
the following year was $82,500. By inference, this property sold at a capitalization rate of
5.5%.
However, we have some subjective information available: (1) Expert opinion is that 5.5%
is “too low” a cap rate for restaurants like this. (2) The subject property is “slightly” riskier,
implying slightly higher than the comparable property. The two signals conform to each
other, suggesting the lower limit of McDavid’s cap rate as 5.5%.
Upper Limit
At the national level (by combining all CRE asset classes together), the cap rate is 7.75%. At
the same time, the capitalization rate for a similar pool of CRE assets in the local market was
9.25%. This information suggests additional CRE risk in the local market compared to the
overall national (as 9.25% > 7.75%). The price of this risk is the spread of 150 bps. Note
that this spread may be different across CRE asset classes. With additional information on
the heterogeneity within the CRE sector, this spread could be further adjusted. However, in
lack of that information, the average spread is the best guess for spread across asset classes.
A similar adjustment could also be made to the cap rate of local restaurants.
We do not have information about the restaurant cap rates in the local area, perhaps
because not many restaurants were transacted in this market recently. However, we do have
information about the national level restaurant cap rate average (i.e., 6.5%). Hence, we could
say that a fair estimate of the capitalization rate of restaurants in the local market should be
150 bps higher than the national average. As a result, our estimate of the average
capitalization rate for restaurants in the local market will be 8%.
Further, we have a piece of subjective information that McDavid’s central locality

promises “substantially” lower risk than overall restaurants in the Denver area. In other
words, the cap rate for the subject property should be substantially lower than 8%.

Page 26 of 93
Fortunately, the signals are, again, in the same direction, making it convenient to assume that
the upper limit of McDavid’s cap rate is 8%.
By the way, the above analysis points towards the quality of communication. Recording
communication cues precisely affords us to differentiate across subjective signals such as
“somewhat,” “substantially,” “definitely,” etc.
By mixing these two sets of information together, here is the information that we have
reached so far: the lower limit for the cap rate for our subject hotel will be five and a half
percent and the upper limit will be 8%. The actual cap rate for our property is expected to
fall in this range. The mid-point of this range is 6.75%.
However, we know from our subjective analysis that the cap rate for our property is only
slightly higher than 5.5% but substantially lower than 8%. That means that it is more likely
that the cap rate will fall on the left-hand side of the midpoint rather than on the midpoint
itself. So, we will go ahead and find the midpoint of the left-hand segment there. The cap
rate estimate will be the average of 5.5% and 6.75% which gives us 6.125%.
A cap rate that is accurate to three decimal places may sometimes be considered to be a
bit “pretentious.” It is advisable to keep cap rates rounded to 25 basis points. Hence our
estimate of the cap rate for this restaurant22 is 6.25%. Remember that the 6.25% is for the
current market. This is our best estimate of the going-in cap rate (𝛾𝑖𝑛 ).
Going-Out Cap Rate
We may also need to estimate the going-out cap rate (𝛾𝑜𝑢𝑡 ). Arguably, in this context, the
business demand (from office employees) for restaurants is less price-sensitive, and hence
less risky than the leisure demand (from households). As a result, the cap rate attributed to
the business demand will be smaller than the cap rate for leisure. The case study suggests
that over time, the demand for this restaurant will shift more towards the leisure segment than
the business demand. If the overall demand profile stays more or less the same, this implies
that in the future, the demand for the restaurant will be riskier and hence 𝛾𝑜𝑢𝑡 will be slightly
higher than the going-in cap rate. At this moment we do not know how much exactly the
price of the shift in the demand will be.
To be conservative but to avoid a drastic change in the cap rate over time, we make an
upward adjustment of a modest 25 bps. Our estimate for the going-out cap rate, in this case,
will be 6.5%.
Appraisal versus Transaction Cap Rates

In an ideal world, the valuation would match the transaction price. In practice, appraised values
are often different from the transaction price. The difference may be due to several factors. The
assumed holding period (that we shall soon discuss) in the market valuation exercise may be
22
The accounting principle of “conservativism” is also adopted widely, although unfairly, by the valuation
community. For a conservative (lower) value estimate, 6.125% is rounded up to 6.25% than to 6.0%.

Page 27 of 93
different from an investor’s investment plan. The NOI, growth rate,

cap rate, and discount rate perception may be different across a
valuer and an investor. Moreover, the transaction may not happen
in a “fair” environment in that one of the parties (buyer or seller) Appraised Value
may have a higher bargaining power than the other. Moreover, the
role of “perception” in valuation may create a deeper schism across versus Transaction
the valuers’ and investors’ value estimates. Behavioral biases in
investors and valuers are well documented in research, and these
Price
biases may go in opposite directions. The end result is that appraisal
The split between appraisal
estimates are taken with a pinch of salt by the asset market. This is
𝑁𝑂𝐼 and transaction-based cap rates
the nature of the business. The appraisal cap rate (𝑎𝑝𝑝𝑟𝑎𝑖𝑠𝑒𝑑𝑡+1𝑣𝑎𝑙𝑢𝑒 ) also penetrates into the
𝑡
𝑁𝑂𝐼𝑡+1 difference between appraised
may differ from the transaction cap rate (𝑇𝑟𝑎𝑛𝑠𝑎𝑐𝑡𝑒𝑑 𝑃𝑟𝑖𝑐𝑒 ). The
𝑡 value and transaction price of
graph below illustrates a scatter plot between the appraisal and assets. A 2016 report by MSCI
transaction-based cap rates recorded across 120 hotel transactions provided several interesting
in the US between 2001 and 2016. The data is sourced from CoStar. insights:
Although the average difference between the two cap rate measures
is relatively small (62 bps), the two cap rates were the same only in • On average, prices were 1-6%
9% cases. In 60% of cases, the difference was substantial (i.e., >100 higher (globally) than the
bps). valuations during 2000-2016.
• On average, the absolute price
Insert Figure 5.10 here difference varied between 9-12%
during this period.
• Valuations were off by nearly 8-
10% of absolute difference in
Some interesting patterns could be observed over time when extreme price segments (lowest,
comparing transaction cap rates to appraisal cap rates. The highest), and by 9-11% in mid-
following graph presents the NCREIF data for the rental apartment price segments.
market in Atlanta, USA: • Absolute differences were
smallest (5-9%) in some
Insert Figure 5.11 here countries such as South Africa,
USA, UK and Italy, and
In recent years, after 2012, some appraisals overestimated,
highest(12% +) in others such as
and some others underestimated the cap rates. The difference has Japan, Germany, Denmark.
been usually within 100 bps. During the crisis, we could see an • The global averages of absolute
over-reaction of appraisal cap rates to the economic scene. For differences were the
example, the appraisal cap rates soared in the quarters immediately highest(around 11%) during the
after the global financial crisis. boom and bust periods
surrounding the global financial
When is the Direct Capitalization Method Applicable? crisis (2006-2009).
Direct capitalization is a relatively simpler method of valuation as
it requires only two inputs to the valuation equation: (1) Projected
NOI from the following year and (2) current year’s cap rate.
However, it is based on a strong assumption that the asset cash
flows have stabilized23. This assumption may be too restrictive in many practical situations. If the
purpose of the valuation is to price a newly developed or redeveloped CRE asset, then the method
23
As discussed earlier, stabilization refers to (broadly) a steady expected growth (g) in cash flows.

Page 28 of 93
is not immediately applicable. Besides, the method is rarely suitable for assets that are to be
transacted. Such assets are considered to experience some periods of unstable cash flows as the
operations mature several periods after an asset was just launched (relaunched) or acquired by a
new owner. There are two distinct uses of the direct capitalization approach as follows:
(1) An estimate of the going-out cap rate can be used to estimate the market value of the asset
in the future after its cash flows have stabilized. This valuation is a critical input to the
DCF method of valuation.
(2) For assets that are already enjoying stabilized cash flows and are being valued for financial
statements, direct capitalization may be a suited method of valuation. For example,
consider Credit Suisse Hotel Funds which owns and operates high-quality CRE assets in
various parts of Switzerland. These assets stabilized several years ago and are expected to
stay on the fund’s balance sheet. The fund needs to keep its shareholders apprised of how
the asset values are moving. A valuation firm could conduct the valuation on a quarterly,
semiannually, or annual frequency by using the direct capitalization method.
Stabilization Period
The period an asset takes to stabilize its operations after being developed/ redeveloped or
purchased is called the “stabilization period”. Stabilization is often observed through occupancy
rates. During this period, the NOIs are less predictable and several estimates are more tentative
than periods when the cash flows have stabilized. During the stabilization period, the cash flows
usually “ramp up” to achieve “performance parity” with other assets. Each geographic market may
have conventional wisdom on how long a certain CRE type will take to stabilize. For example, in
most markets in the US, luxury hotels (3.3 years) take longer than relatively lower quality
“upscale” hotels (2.9 years) to stabilize24. The stabilization period refers to the fit between an asset
and its demand in a locality. Therefore, one cannot have universal rules for stabilization periods.
For example, it is believed that the “ramp up” period for luxury assets may be even longer in
emerging markets, such as India or China. Nevertheless, it is expected that most assets will
stabilize within five years.
Discounted Cash Flow Method

Knowing a bit more about cap rates will help us progress further into a second, but more
sophisticated Income Approach to the Valuation method called the discounted cash flow
(DCF) method. We shall see that the DCF method depends heavily on the Direct Capitalization
method.
The Big Picture
As a CRE asset is associated with a specific projected cash flow schedule, it can be valued using
the income approach. In most cases related to asset transactions, new development, or extreme
market cycles (such as DotCom bust of 2001, Global Financial Crisis of 2007, Covid19 Crisis of
2020, etc.), the Direct Capitalization method is rendered irrelevant due to a lack of confidence in
24
See O’Neill (2011)

Page 29 of 93
the stabilization of cash flows. In such scenarios, it helps to be more realistic about cash flows
projections of the next few years. A valuer assumes that after a few years, the cash flows will
broadly stabilize.
As a big picture understanding of the DCF valuation method, we list the major steps of
analysis as follows:
Step-1: Decide on a holding period (HP) for valuation analysis.
Step-2: Conduct a thorough market analysis to develop assumptions about CFO.
Step-3: Develop assumptions about valuation metrics: 𝛾𝑖𝑛 , 𝛾𝑜𝑢𝑡, cash flow growth rate, Ω,
cost of purchase and sale, etc.
Step-4: Forecast Cash flows.
Step 5: Calculate the GPV.
In the following sections, we shall discuss these steps in detail.
Holding Periods
By assuming a finite holding period, a valuer reduces the virtually infinite operational life of a
CRE asset to a manageable time interval (= 𝐻𝑃) for which cash flows could be confidently
forecasted. A necessary condition for the assumed holding period is that it must be larger than the
stabilization period. This condition is necessary to assume that the sale price of the asset (at the
end of the holding period) may be estimated using the Direct Capitalization method.
In selecting a holding period for valuation analysis, a valuer is deciding on a tradeoff:

Longer HPs may provide a more realistic picture of valuation by mitigating the effects of market
cycles; but it requires additional effort in detailed cash flow projections for each extra year added
to HP. Scientifically, it will be impractical to come up with precise forecasts of cash flows too far
out in the future. As a result, using a five- or 10-year holding period has evolved as an industry
norm. The choice between five or 10-years, too, is not without some additional considerations, as
we shall soon review (i.e., smaller holding periods require an adjustment to the discount rate Ω.)
We must not confuse the valuation of HP with the actual HP planned by an owner. The HP of CRE
may vary substantially with location, asset type, and owners. The following exhibit plots the
probabilities (Y-axis) of different HPs (X-axis) based on a sample of over 100,000 CRE assets
owned by US firms. Nearly 50% of all CRE assets are sold within the first 10 years. By the 20th
year, nearly 80% of assets are sold.
The following exhibit breaks up the data by main property types. We can observe
significant variations across property types: some asset classes, such as office, rental apartments
(“multifamily”), and industrial assets, tend to have shorter HP, as most of them are sold within the
first 10 years.

Page 30 of 93
HPs may significantly vary within the subclasses of an asset class, as shown with an
example of hotels below:

Besides, research suggests25 that holding periods are a complex function of owner type,
capital expenditures, and market conditions. It is difficult to predict the actual HP of an asset, and
hence it may be practically a futile exercise to match the valuation of HP with the actual HP of an
asset.
Discount Rate Estimation

We have elaborated earlier in this chapter on how to determine various cap rates. In particular, we
are interested in 𝛾𝑖𝑛 and 𝛾𝑜𝑢𝑡 for a specific asset after all necessary adjustments, including the
adjustment for capital reserves. Estimation of 𝛾𝑖𝑛 , in particular, helps estimate the discount rate Ω.
Mathematically, Ω = 𝑦 + 𝑔, where 𝑔 is the stabilized growth rate in cash flows. Given an estimate
of the cap rate (𝑦), the discount rate will depend on the growth rate promised by an asset. Ideally,
one should first develop a cash flow forecast for the assumed holding period wherein the NOI is a
result of numerous revenue and expenditure items that may follow different growth regimes. In
the long run, the growth rate in these items should converge to the same number. An analyst’s
endeavor should be to estimate the stabilized growth rate during the final years of the analysis, as
presented in the Pro-forma26. By adding this growth rate to the cap rate, one estimates the discount
rate for valuation.
Some analysts argue that similar to cap rate, for a market value of an asset, the Ω must also
be derived from market participants’ perception of the discount rate. Therefore, the valuation
community takes a similar approach to estimating Ω as to estimating the cap rate. Some market
reports may offer the discount rate estimate in a similar fashion. This approach requires an ex-ante
(before the fact) forecast for the growth rate as well. However, such an approach may lead to the
“growth rate dilemma” wherein the ex-ante growth rate estimate does not match with the Pro-
forma growth rate. In the Appendix to this chapter, we discuss some methods applied to estimate
growth rate as such and the resulting growth-rate dilemma.
Cash Flow Forecasts
Forecasting cash flows is a critical component of CRE valuation and depend on numerous
assumptions related to real estate operations. Our main focus in this chapter is on assets that are
developed or acquired with a view to generate rental cash flows. Our examples include assets that
promise sustainable rental income without substantial repositioning. However, the principles
learned here may also be applied to assets that are renovated during the holding period.
“It is difficult to make predictions, especially about the future”

-Neils Bohr
25
See Poretti & Das (2020)
26
“Pro-forma” in Latin means “for formality”. However, in investment circles, it has a specific connotation: A detailed
account, preferably in spreadsheet format, of future projections of cash flows, risk metrics and decision criteria.

Page 31 of 93
The Pro-forma
One of the central aims of Pro-forma is to lucidly present the forecasts of NOI of the asset during
the holding period (and a year beyond). However, forecasting the NOI directly is a difficult
endeavor, especially during the early years i.e., the stabilization period. A valuer breaks the
challenges up into different components of revenues, expenses, and reserves. During the holding
period, it is expected that the valuer will pay careful attention to details and come up with plausible
forecasts for each line item individually, per period. Forecasting the line items requires a great deal
of market study and subjective judgments. If a valuer is lucky, she has access to past cash flows
statements of a hotel. In some cases (e.g., hotels), some consulting firms may also provide
“benchmarking” reports that include “profitability reports”. These reports collect data from
multiple assets and standardize them (e.g. into per room or per square meter basis). Thus, a
benchmark profitability report represents cash flows that a “typical” asset of a given set of
characteristics (sub-segment, location, etc.) has generated in the past.
A marginal investor in the asset could address the suboptimality in operations by altering
the management philosophy. The new management may not be willing to pay an extra price for
the extraordinary operational performance of the existing management with an argument that it
may be unrealistic in a long run. If the aim is to develop a market value estimate, the market
benchmarks could be more relevant than the idiosyncratic operational characteristics of the
existing asset (i.e., its own past statements of cash flow). Yet, past cash flow statements can help
identify sources of revenues and expenses specific to the asset. Besides, suitable market
benchmarks are a luxury in most markets. Therefore, for existing assets, it may be pragmatic to
review the past cash flow statements to forecast the future. Anomalous items in the revenue or
expense sides (measures that do not conform to perceived market norms) may need to be corrected.
Some extra-ordinary considerations (e.g., presence or lack of important amenities) must be
accounted for before developing the forecasts. If the asset is brand new, then careful consideration
is warranted in listing and quantifying different line items in the revenues and expenses category.
In our example, the cash flow proforma will be depicted in tabular format. Each row will
list a specific line item of the cash flow statement (e.g., revenue or expense). Each column will
represent a future year for which the cash flow is being forecast. Recall the VastraMart example.
At the end of this chapter, we provide a detailed case study of “Hotel Himalaya Imagine.” For
the time being, in an endeavor to appreciate the fundamentals, we shall continue with our simple
example of VastraMart.
Revenues
Rents are the most important component of a CRE cash flow statement. The rental rates are tracked
by professionals and consultants. As a result, markets have conventional wisdom on ongoing rental
rates. Else, talking to local brokers or surveying potential tenants may offer a good sense of the
possible rental rates. There may be different tiers of rents within a facility. For example, some
office suites may be pricier than some others. In such cases, the rents could be split into multiple
rows corresponding to each category. The following figure 5.15 shows the evolution of the office,
retail, and industrial rental rates in some leading CRE markets in India. It is evident that these rates
vary significantly across locations and asset types:

Page 32 of 93
In some markets, one may have access to forecast of rental growth rates. Else, forecasts of
GDP growth, local inflation rates, and new competitive supply may be subjectively combined to
build an educated assumption of growth in rental rates. Note that CRE leases may be signed over
longer terms. Thus, the contractual rent escalation rates (and subjective probabilities of lease
renewal or termination) may lead to irregular growth patterns. It may not be sustainable, however,
to assume a long-term rental growth rate to be too different from inflation rates. For the first few
years, a realistic assumption on a “ramp up” in rental rates may be dictated by the marketing
strategy.
In addition to rental revenues, a CRE facility may have other revenues (e.g., restaurant,
spa, parking, etc.). A separate line item must be created for all revenue sources. Their growth rates
may also be subject to local nuances, but should broadly move along the consumer price index
unless during the stabilization period. Depending on the specificities of the asset, these revenues
may be closely tied to the rental revenue or may evolve separately. A valuer must discretely assign
a revenue growth pattern to each revenue item.
Rental rate per unit (e.g., per square meter, per suite, per room, etc.) multiplied by the number of
units, in conjunction with other revenues provides an estimate of PGI.
Vacancy loss
As discussed earlier in Chapter 2, some vacancy (i.e., percent of the rentable area that remains
unrented) in most CRE assets is essential and desirable. If a market forecast is available, one should
adapt the vacancy rate forecasts accordingly. Else, a careful look at the lease contracts may suggest
periods of vacancy caused by an outgoing tenant. An overly optimistic estimate of (low) vacancy
rates compared to the market average may warrant excess expenses allocated towards sales &
marketing, as well as lower rental rates. In larger assets, this could be an ongoing activity, such
that a portion of the asset remains dynamically but perpetually vacant. For future years in the
holding period, it is advisable to consider the market cycle in mind. As such, the vacancy rates
may sporadically vary even during periods of stabilization. Besides, a prudent owner must
provision for additional vacancy once every few years (e.g., 7 -10) for capital improvements. While
a portion is being improved, it must remain vacant.
Potential gross (rental) revenue multiplied by the vacancy rate results in vacancy loss
(VAC). PGI (inclusive of all revenues) minus VAC equals effective gross income (EGI).
Variable Expenses
Unless an asset warrants some specific considerations, the forecast for expenses is relatively
simpler. Variable Expenses depend on revenues. Local market experts, market benchmarking
reports, or facility managers may be consulted to estimate the expense ratio (i.e., expense/ revenue)
across different rental incomes. Some rental segments may have higher expense ratios than others.
However, their combination may be necessary to build the synergy effect. For example, although
restaurants in a facility may have a high expense ratio (i.e., low profitability), it may still be nice
to keep them in the facility to “lure” tenants from other areas. As these variable expenses are
directly proportional to revenues, one need not estimate growth rates separately for them.

Page 33 of 93
Fixed Expenses
Some other expenses may not be tied to revenues. For example, an owner may incur General &
Administrative (G&A) expenses, common area management (CAM) expenses, property taxes,
insurance, facility operator fees, etc. irrespective of variations in revenues. A valuer estimates these
expenses according to market research and allocates their year-over-year (YoY) escalation
(growth) rates separately. In most cases, these escalations are close to the inflation rate.
Hybrid Revenues and Expenses
Sometimes, some revenues are partially tied to rental revenues. For example, in a luxury hotel, the
spa and restaurants may attract local customers, in addition to the hotel room patrons. Such
revenues and expenses could be divided into two sub-items. The variable component could be tied
to rental revenues. The rest could be allowed to evolve with their appropriate escalation rates.
Property Taxes
In common law countries, property taxes are levied “ad valorem” (i.e., as a proportion of the value
of the asset). As such, incorporating property taxes in the cash flow Pro-forma for valuation leads
to the problem of “circular referencing.” In other words, property tax estimates are required to
estimate the value, as well as vice versa. The good news is that, although loosely tied to the market
value, property taxes broadly depend on how the local government assesses the taxable value of
an asset. In many cases, these taxes are detached from realistic value estimates of assets.
Governments may prescribe fixed per-unit area (e.g., floor area-based) tax rates. A valuer must
survey the local regulation to estimate the property tax rate for the first year, and then assume an
appropriate YoY escalation rate.
The sum of all the expenses is the OPEX
EGI minus OPEX equals asset-level EBITDA.
Capital Reserves
Usually, a certain percent of total revenues is kept aside in the name of reserves for replacement.
Most owners would allocate 2-5% of their revenues to this budget annually to take care of FF&E
replacements, TI, LC, etc. Such a regular budget for improvements should not materially alter the
rental rates or occupancy in the following years.
However, it may be more optimal for some assets to go through a major renovation,
redevelopment, or addition to their facilities during the holding period. A good valuation analysis
will infuse this information in the proforma by allocating a larger budget in capital reserves (which
may be distributed over several years or maybe consolidated in bigger chunks within a smaller
period).
There may be uncertainty about when exactly the capital improvement takes place.
Consulting with existing (or prospective) owners may provide some sense of the most suitable
year (during the holding period) for the improvements. It is that particular year for which the
occupancy rates must be adjusted down. However, this additional budget deserves to be
compensated with superior performance (rental rates, or occupancy rates) in the later years.

Page 34 of 93
In some special cases, an existing asset may require a substantial budget for capital
improvements at the outset just for the asset to achieve the benchmark performance level. A capital
improvement plan will dictate how long the improvements will take place. Usually, the first few
quarters should be sufficient. The GPV of the capital improvement budget should be deducted
from the “full potential” valuation of the asset to estimate its market value27.
EBITDA minus capital reserves equal NOI.
VastraMart: DCF Valuation

Let us build our proforma of the VastraMart case further. The 2018 cash flows are as discussed
earlier. The valuation is conducted at the end of 2018. For simplicity, let us assume that the asset
has stabilized from the outset and there are no fluctuations in the occupancy rate which remains
stable at 90% (i.e., vacancy rate = 10%). Consider a five-year holding period, and 4% growth in
PGI, MI, and OPEX. Assume that the market going-in cap rate (𝛾𝑖𝑛 after all adjustments) is 10%,
𝛾𝑜𝑢𝑡 = 10.5%. The cost of sale is 4%. The resulting cash flow proforma is illustrated below:
Insert Table 5.3here
Considering a five-year holding period warrants projecting future cash flows for six years
(i.e. 2019 through 2024). The 2024 NOI is capitalized to the sale price using the going-out cap
rate. 4% of the sale price goes towards the cost of sale. The difference is the NSP. The final year
asset cash flow is the sum of NSP and NOI. For earlier years, the cash flows equal the NOI. From
the Pro-forma, one can observe that the y-o-y NOI growth rate (g) has stabilized at around 4%
The discount rate (Ω) equals 𝛾𝑖𝑛 + 𝑔 = 14%. The GPV of future Asset Cash flows is the estimated
market value of the asset. Note that all revenues and expenses are growing at the same rate (4%)
which is also equal to the ex-ante growth rate. As a result, the ex-post growth rate in NOI is also
exactly equal to 4%.
How does the DCF value of VastraMart compare to the Direct Capitalization value
estimated earlier? What could explain the anomaly? This is the next topic of discussion.
Some Practical Issues with DCF Valuation

Holding Period Assumption
In the VastraMart case, the CRE asset has stabilized at the outset. The growth rate in revenues,
escalation rate in expenses, and expense ratios are fixed and the same. We should expect to have
a definitive view on its valuation given the cap rate, growth rate, and discount rate. Yet, the Direct
Capitalization method led to a different (higher) value estimate than the DCF method. What would
happen if we considered a 10-year proforma for the same asset? As shown below, a differently
assumed holding period will provide yet another value estimate.
27
Note that capital improvements are considered as investments, and may not be eligible as an Expense” for tax
deductions immediately. However, an owner can depreciate it in the future years to enjoy the tax benefits. We discuss
the taxation later in Chapter 7.

Page 35 of 93
The sensitivity of CRE valuation to the assumed holding period perplexes some students
of general finance. However, this is a result of incorporating practicalities, although imperfectly,
into valuation. CRE valuers are sensitive to the effect of ageing on the asset value. As a result,
valuers assume a spread ∆ between the Going-out and Going-in cap rates (= 𝛾𝑜𝑢𝑡 − 𝛾𝑖𝑛 ). We do
not observe a similar adjustment in, say, stock valuation wherein a stabilized series of cash flows
will be independent of the assumed holding period28. The following exhibit shows how the
valuation of a similar asset would vary over differently assumed holding periods. The X-axis
depicts the assumed holding period (in years). The solid graph depicts the corresponding value
(left vertical axis). The dotted line shows the deviation from direct capitalization value along the
secondary (right) vertical axis29.
The deviations are particularly severe when the assumed holding periods are small.
Therefore, it is advisable to go for a longer holding period in valuation endeavors. It should also
be safe to assume that the cap rate spreads (∆= 𝑦𝑜𝑢𝑡 − 𝑦𝑖𝑛 ) refers to a typical holding period of 10-
years. For smaller holding periods the spread may be very small or negligible.
Multitude of Value Estimates

After much work on a valuation exercise, a valuer would wish that the asset is transacted at the
appraised value. As we have seen earlier, that is rarely the case. A DCF valuation exercise is
strongly susceptible to the “Garbage-in-Garbage-Out” (GIGO) phenomenon. If the assumptions
for valuation are inaccurate, the valuation will be inaccurate. Thus, one is uncertain about the
match of the appraised value to the realized market value (from a transaction). Besides the GIGO
phenomenon, one must acknowledge the probabilistic nature of most assumptions: revenues,
expenses, cap rate, growth rate, discount rate, and others. As a result, even if the assumptions are
not prone to errors and systematic biases, the resulting valuation is one instance from numerous
possible value estimates.
A good valuation exercise acknowledges the probabilistic nature of valuation. The issue
can be addressed in numerous ways. We are listing some in their order of increasing complexity.
Scenario Analysis
A valuer may run several “scenarios” of market conditions. The scenarios are with respect to the
appraised value in that an optimistic scenario will lead to higher valuation, and a pessimistic
scenario will lead to a lower valuation. The baseline scenario may be considered as the “most
likely scenario.” In the “pessimistic scenario”, revenues are at their lowest, whereas expenses, cap
rate, and discount rate are at their highest possible values. In the “optimistic” scenario, revenues
are at their highest, whereas expenses, cap rate, and discount rate are at their lowest possible values.
28
Recall that ∆= ∆𝛾 + ∆𝑎𝑔𝑒𝑖𝑛𝑔 . We should also expect the ∆𝛾 to vary across the HP (i.e. market cap rates will evolve
over time). In the current discussion, for simplicity we assume that the market cap rates are stable (i.e. ∆𝛾 = 0) and all
the ∆ is purely due to ageing.
29
The Going-In Capitalization Rate (𝛾𝑖𝑛 ), Going-out capitalization rate (𝛾𝑜𝑢𝑡 ), NOI growth rate (𝑔) and the first-
year cash flow are assumed to be respectively 10%, 11.5%, 5% and $1.25 mi respectively. It is assumed that the cash
flows are stabilized from the outset such that the true value of the asset equals $1.25 mi / Yin = $125 mi.

Page 36 of 93
The assumptions in these scenarios must still be realistic and broadly in line with their observed
ranges in the market.
Sensitivity Analysis
Alternatively, a valuer examines the sensitivity of the valuation to changes in one variable at a
time. Keeping other assumptions unaltered, the valuer examines the resulting value estimates
across different possible values of a selected variable. Then, one plots a scatter plot of the data
wherein Y-axis depicts the valuation, and the X-axis depicts different values of the assumed
variable. The steeper the curve, the more sensitive valuation is to the selected variable. The
following example depicts the sensitivity of appraised value (in a specific asset) with respect to
several assumptions.
In this rather typical example, the discount rate curve is steeper than the cap rate, implying
the importance of the growth rate assumption. Similar sensitivities could examine across other
assumptions, such as rental growth rate, expense ratios, etc. Assumptions with steep curves warrant
a careful analysis in developing them. Sometimes, valuers see the sensitivity of appraised value by
letting two assumptions change simultaneously, as shown below. The result can be shown as a
table. The two axes represent respective assumptions and the values in the middle are the
corresponding appraised value. In Excel, the “WHAT-IF” analysis functionality could help
automate this exercise. An example is shown below.
Monte Carlo Simulations
The methods shown above break continuous probabilities into discrete values of assumptions. A
more sophisticated method would appreciate that the probability distribution of each assumption
may be continuous. Past values of variables may dictate the nature of the probability distribution.
Else, it is pragmatic to assume a “normal” distribution for the selected variable for which the mean
and standard deviations are known. For a simple illustration of Monte-Carlo analysis, let us revisit
the VastraMart Case again.
Assuming a five-year holding period. We include our selected variables (i.e., adjusted
discount rate and annual cash flows) in the simulation. Each row presents one of the virtually
infinite probabilistic scenarios30. The first row of the table shown below (“Mean”) depicts the most
likely value of each variable derived from our “baseline” scenario analyzed earlier. Arguably, these
values are the average from the probability distribution of respective variables.
The second row is our best guess of the minimum value for each variable. This is not to be
confused with the “worst-case scenario” wherein the discount rate should have been at its highest
probable value. The third row depicts the “Max” scenario, where each variable is at its highest
probable value.
30
In this case, a scenario is a specific set of our selected variables.

Page 37 of 93
The highest and the lowest values may need to be subjectively assigned, but must be
dictated by the realities of the market. We could assume that 90% of the valuations will fall within
our range of minimum and maximum values, which is equivalent to four times the standard
deviation of the distribution. Accordingly, we can estimate the standard deviation by dividing the
range by four. These estimates are shown in the last row of the table. By identifying the mean and
standard deviation of a variable, we are defining a unique normal distribution for it.
The next step would be to generate numerous (usually hundreds or thousands) probable
scenarios. In a scenario (row), each variable is drawn randomly from its probability distribution.
Spreadsheet applications come in handy when conducting this exercise31. We also create an
additional column of the resulting value (i.e. GPV based on simulated cash flows). A snapshot of
five thousand such scenarios created is depicted below:
A visual analysis of the 5000 appraised values is presented in the frequency diagram below.
The average of simulated values is ₹144,884,606 with a standard deviation of roughly ₹1.1 crores.
Therefore 95% confidence interval of the expected valuations will be roughly ₹14.5± 2 ×
₹1.1
crore i.e. we are 95% confident that the margin of error in the expected value estimate is
√5000
±₹0.02 crore.
If all variables are normally distributed, we should expect that the mean of all simulated
values will be close to the baseline valuation. However, the range of possible values in the 90%
confidence interval is valuable information, as it shows the riskiness in valuation. Two possible
assets may have the same expected valuation, but the one with lower risk will vary within a
narrower range. From an investment valuation (IV) standpoint, this information could be useful to
adjust the risk premium in expected return.
Majority of our discussion so far has focused on the income approach to valuation.
However, one expects a valuation report to provide value estimates from different perspectives
and using different methods. In the following sections, we provide a brief overview of some other
methods.
Cost Approach
The Cost Approach to valuation is based on Adam Smith’s32 proposition that it costs to produce
something; this cost is its “natural price. Menger disagreed. He stressed on the role of demand in
valuing something: Something may be highly valued because of its demand even if the cost to
produce it is much smaller. Yet, well-functioning markets may push the cost of raw materials (e.g.,
land, steel, etc.) up for assets that are in high demand. Therefore, a valuation estimate from the
Cost Approach may not necessarily be the same as the market value, but it may be positively
correlated with it.
31
Explore the NORM.INV() function in MS Excel that generates a random number from a specified normal
distribution.
32
The Wealth of Nations (1776),

Page 38 of 93
Traditional valuation practices in India had been dominated by the Cost Approach, where
a civil engineer or architect were considered the most suited for valuing real estate. This tendency
is still prevalent in the residential sector, where it is less intuitive to apply the income approach to
valuation. However, while inputs from engineering and architectural viewpoints may be a
necessary ingredient of the Cost Approach, it is definitely not sufficient in most cases for two main
reasons: 1) beyond the cost of development, a valuer must also assign some premium and discount
to bring the cost close to market valuation 2) The subjectivity involved in this process, especially
in the case of commercial real estate, warrants a good grasp on market forces, both in terms of real
estate and capital markets. Further, depending on the market forces, the role of depreciation may
be critical in certain valuation projects.
Consider a luxury hotel built on a freehold land parcel that has signed a management contract
with a global hotel chain. The Cost Approach divides the valuation into three components:
1. Land Cost: The Cost of acquiring the land parcel in the current market conditions, as if
nothing were built on it yet. Valuers usually apply the Sales Comparison Approach
(discussed in the next section) to estimate the cost of acquiring the land parcel.
2. Hard Cost: The cost of installing the tangible components such as building, FF&E,
infrastructure, etc. are estimated through engineering/ architectural survey, or simply by
reviewing a cost index.
3. Soft Cost: For an asset to be completely developed, there may be costs related to intangible
products, services, and contracts. Fees to be paid to architects, engineers, consultants, and
lawyers may be necessary to develop an asset of similar quality. Besides, an asset may
include other intangible assets, such as a brand franchise, management contract with third
parties.
The sum of land, hard and soft costs leads to an estimated value of the asset as if it were brand
new. The asset in question, however, may be old. A valuation norm is to apply a depreciation due
to ageing. Notably, depreciation only applies to hard and soft costs. For simplicity, many valuers
apply a blanket depreciation (e.g. 20%) to all the depreciable assets. However, this depreciation is
not the same as accounting depreciation. Again, a valuer must survey comparable assets to assess
the percent reduction in value due to ageing and apply a matching depreciation rate to the subject
asset.
While it may be relatively simpler to estimate the various costs, applying an appropriate
depreciation is a conundrum and is subject to massive errors. It may be challenging to apply a
fixed schedule for economic depreciation in an asset: the depreciation may vary by asset class,
location, and time. Besides, some assets may actually enjoy an appreciation after development due
to increased recognition, market acceptance, or, simply, improved operational efficiency over
time. The Cost Approach may be blind towards these possibilities and may force a reduction in
value due to ageing, which may not be accurate.

Page 39 of 93
Two luxury hotels in Chicago Downtown promise similar amenities and are of broadly the same
size. One is 50 years old. Another is 90 years old. Age is the only material factor that could
cause the two hotels to have different valuations. Which hotel will sell at a higher price?
A recent study33 has shown that for about the first several (say, 60) years, US hotels lose
value due to ageing. 2% of the remaining value is written down annually. Afterwards, the values
start to appreciate owing to the “vintage effect”, again at 2% written up every year. As such, the
hotel will lose value between 50 and 60 years of age but will gain even more value between the
60th and 90th year of age. The result will be that the older hotel will rather enjoy a price premium.
Note that although the vintage effect may be applicable in most parts of the world, the point
of inflection (i.e., the age after which being old is valued more) may be different across
geographic markets. In older urban civilization (e.g., Western Europe), where a neighborhood
may be abundant with early 20th century buildings, the “vintage effect” may not be applicable
to most buildings. In the US, however, the vintage effect may kick in earlier.
Similarly, the vintage effect in a New Delhi hotel that is housed in a (well-functioning and
maintained) 100-year-old building may be much stronger and positive compared to a similarly
sized hotel housed in a Haveli of Sikar (Rajasthan) where being an old haveli is no rarity.
To assess the suitability of the Cost approach in valuation, here are some relevant aspects to
examine:
1. Is the asset too old that renders the cost approach irrelevant?
2. Was the sales comparison approach for the land valuation conducted appropriately?
3. Do the development costs include an acceptable estimate? Are all the hard and soft costs
estimated carefully?
4. Is depreciation a real concern? Rather than a straight line, or written down depreciation,
could some other methods of depreciation be more accurate?
Example: POTUS Tower, a high-end, branded serviced apartment building of 1,200 sqm
located in Safdarjung Enclave, New Delhi is being listed for sale. The three-story building has
a floor plate of 400 sqm. The facility includes 16 fully furnished studios, a basement parking
lot of a 400 sqm floor area, and a security/maintenance office included on the ground floor.
The building stands on a 600 sqm plot of land. The building is 10-years old.
Land Cost: In this locality, land parcels are selling at a rate of ₹500,000 per sqm. Therefore,
the cost of land will be 600 × 500,000 = ₹30 𝑐𝑟.
Hard Cost: The hard cost has three main components as follows:
33
Das, P., Smith, P., & Gallimore, P. (2018). Pricing extreme attributes in commercial real estate: The case of hotel
transactions. The Journal of Real Estate Finance and Economics, 57(2), 264-296.

Page 40 of 93
Civil work: The cost of constructing the superstructure is ₹30,000 per sqm. The cost of
basement construction is 50% higher. Therefore, the cost of civil works will be
1200 × 30,000 + 400 × 30,000 × 1.5 = ₹5.4 𝑐𝑟.
MEP34 work: Other engineering works (including the elevator) will cost around ₹200 per
sqm. The total MEP cost will be 200 × (1,200 + 400) = ₹0.32 𝑐𝑟.
Elevator and Security Systems: The elevator, security system, and other similar
installations will cost around ₹ 1cr.
FF&E: The FF&E in the rooms will be around 20% of the civil work equaling
₹5.4 𝑐𝑟.× 25% = 𝑅𝑠. 1.35 𝑐𝑟.
Total hard costs are ₹8.07 cr.
Soft Cost: the combined fees of architects, landscape designers, interior designers, and
engineers is estimated to be around ₹0.25 cr. The signing fee of the residential brand was
around Rs. 2 cr. (for a 20-year contract). This contract was signed 10 years ago. Hence, the
remaining amortized value is Rs. 1 cr. Besides, there may be some costs involved related to
legal work and various liaisons. They amount to ₹1 cr.
Total soft costs will be ₹2.25 cr.
The sum of the three types of costs is ₹40.32 cr. This is equivalent to the value of a brand new
asset of the given characteristics.
Depreciation: You do not have close comparables available. However, strata-title high-end
apartments of similar age are selling at a 5% discount35. Therefore, the depreciation applicable
will be 5% of the total cost excluding depreciation.
Final value estimate: 40.32 − 8.07 × 5% ≈ ₹40 𝑐𝑟.
34
MEP stands for mechanical, electrical, and plumbing
35
Many valuers would depreciate all the non-land components as a single item. As this method is, anyway, a broad
approximation, and the error infused by including or excluding some items is minimal, in this case those nuances are
not a big concern.

Page 41 of 93
A Side Note:
Suppose your fund acquires this property at. Rs. 40 cr. Considering the suite rentals as the only
source of revenue, and 10% towards all expenses (including capital reserves), how will you
estimate an optimal rental rate for these properties?
A viable solution would be to start from the market capitalization rate. Suppose the market
cap rate for such properties is 6%.
Projected NOI for the next year = 6% × 𝑅𝑠. 40𝑐𝑟. = ₹2.4 𝑐𝑟.
Rental Revenue *(1-10%) = ₹2.4 cr.
Therefore, Rental Revenue = ₹2.67 𝑐𝑟.
→Annual rent per suite = ₹2.67 𝑐𝑟./16 = ₹16.7 𝑙𝑎𝑘ℎ

→Monthly rent per suite = ₹16.7 𝑙𝑎𝑘ℎ/12 ≈ ₹1.4 𝑙𝑎𝑘ℎ
Now, consider 20% vacancy on average. This would increase the expected monthly rent per
suite to
₹1.4 𝑙𝑎𝑘ℎ/80% = ₹1.7 𝑙𝑎𝑘ℎ𝑠.
These suites are going to be expensive after all! What would be your solution to make these
rentals more affordable? Hint: the FSI could be maximized up to 3.5.
Sales Comparison Approach

In CRE markets, the Sales Comparison approach is mostly a validation tool for the DCF valuation.
This method is relatively more popular in residential valuation. Unlike the Income or Cost
approaches to valuation, the Sales Comparison approach is less fundamental: It does not promise
to estimate the “intrinsic” or “natural” value of an asset. Rather, it is a more pragmatic approach
focused on directly guessing the price at which an asset with certain characteristics will sell in a
market. The central idea behind this approach is to record the pattern in which investors have
priced assets of similar qualities in the past and extrapolate those simple patterns to valuing a
subject asset.
“[Sales Comparison] … just prevents you from being stupider than all other active buyers”36
While the Sales Comparison Approach is usually treated as a “back-of-the-envelope” or

“thumb rule” method to valuing assets, one may still adopt a slightly more scientific and
sophisticated approach while valuing an asset while adopting this approach.
In developed countries availability of data on comparable sales is not as big an issue as in other
countries such as India. Indian real estate markets suffer from an acute shortage of commercial
real estate data whereas US markets have large data companies, some of which are listed on stock
36
Cook, S. (2018). Investing in Real Estate Private Equity : An Insider’s Guide to Real Estate Partnerships, Funds,
Joint Ventures & Crowdfunding. Kindle Independent.

Page 42 of 93
exchanges. However, a valuer needs to have multiple sources of transactions data, even in the
developed countries, as no one source guarantees to be comprehensive. The Appraisal Institute
(USA) cautions against collecting data from “someone who is not a party to the transaction
provides sales data because the motivation of the parties to the transaction is an important
consideration. Sometimes brokers will be able to provide more reliable information than the buyer
or seller.” Before being included as comparable sales, a transaction must meet a few conditions.
In other words, a transaction price can provide per unit fair market value if the conditions are met:
1. The source of data is reliable.
2. The asset was sold recently and within the same phase of the market cycle. Else, necessary
adjustments must be made to bring the past transaction price at par with the current time.
3. The asset belongs to the same CRE segment.
4. The asset is in the same or similar locality.
5. The transaction happened at arm’s length37
6. The sale happened in normal conditions (it was not a foreclosure sale). Else, necessary
adjustments may be warranted.
Data is the New Oil
Proprietary databases are increasingly expensive, restricting their access to those who can
afford it. As a result, several consultants tend to maintain their internal database. The data
may be sourced from newspapers, other media, or street talk. It is advisable for all real CRE
enthusiasts to maintain their personal database and share it with each other. It is also
important to collect data in a structured format so that searching for specific information
based on certain criteria will be easy in the future. Simple tools such as Google Forms could
help in building a good structure for data.
Error Inbreeding
In Sales Comparison, self-collected data can be handy as it may fit better with the local
context. Yet, one must be careful in verifying the self-collected or self-generated (e.g., past
valuations) data. The Market value data from appraisal must be reviewed with caution.
Incorporating a past appraised value into a Sales Comparison exercise may continue to bias
a valuer’s estimates if the original valuation was erroneous or biased.
Valuers must endeavor to keep their valuation as objective as possible.
Sales Comparison by Per Unit Price

As one of the simplest methods, a valuer collects the sale price information of comparable assets
that were recently transacted on a per-unit basis. If some of the transactions are relatively old, a
valuer could adjust the value for comparability with the current times by applying an appropriate
percent change since the transaction date. Property price indices may provide useful clues. The
units of comparison could vary. Price per square foot (or square meter), as well as Gross Revenue
Multiplier (GRM), are applicable almost universally. However, several other units may make more
sense.
37
A transaction when the two parties are independent of (unrelated to) each other and acting in their best interests.

Page 43 of 93
Example: During early 2019, you needed to value a 1500 sqm warehouse in Greater Noida,
a suburb of the National Capital Region (NCR). Three industrial warehouses of similar
quality were sold in the suburban NCR region in late 2018. Their prices ranged between
₹50,000 and ₹85,000 /sqm, averaging at ₹70,000 per sqm.
The subject asset could be priced at 1500 × 70,000 = ₹10.5 𝑐𝑟𝑜𝑟𝑒s.
The above example is based on a strong assumption that the comparable properties are very
close in characteristics to the subject asset. In practice, access to such data is a luxury.
Example: Sales Comparison by Gross Revenue Multiplier

Recently, 4 offices were sold in the Ahmedabad area whose data were collected as follows:
The market has been stable in recent years and is expected to remain so in the near future.
You need to value an office whose estimated gross rent is currently Rs. 6 lakh per year.
Although it may be difficult to collect information on the NOI of comparable
properties, estimating the gross rent may be a relatively simpler task. So long as all
comparable properties have similar operational efficiency (i.e., expense ratios, occupancy
rates, etc.), the gross revenue multiplier (GRM) calculated as the ratio of gross annual rent
over asset price may be a handy valuation ratio.
In real-life scenarios, the GRMs may be widely dispersed and may be collected
from sales that happened at different phases of the market cycle. A valuer may have to
subjectively exclude some outlier observations before calculating the average. In this
example, considering the average GRM, the value estimate will be:
₹ 600,000 × 13.19 ≈ ₹ 79 𝑙𝑎𝑘ℎ
Example 3: Adjusted Sale Price

Sometimes, the comparable properties may differ from the subject property based on
specific characteristics that may impact their valuation. First, one collects price information
of comparable properties. The next step involves listing the criteria across which the assets
are different from the subject asset. We also need a method to quantify the differences.
When exact measurements are not available, we could use a binary measure (absent versus
present; yes versus no).
Consider this example, where the subject property has 300 keys. Its valuation
criteria are listed. The subject property has a freehold land title (not a ground lease), it is in
good condition (but could be better), etc. Let's start with Comparable One. The comparable
property was recently sold for $73 million, but it was a 250-room property only. On a per
room basis, the transaction price of this hotel was $73 million, implying $292,000 per

Page 44 of 93
room. In a simple Sales-Comparison approach, we would calculate the price per key across
different comparable assets. The average price per key would be the “pricing pattern” and
applicable to the subject property.
But in this method, we acknowledge that our comparable sales are not exactly the
same as our subject property. We try to price the differences that we have identified. For
example, the subject property is supposed to be in good condition, but the comparable
property is in excellent condition. This condition sounds quite subjective, but in the real
world, as a valuer gains experience and expertise, she might be able to estimate the capital
expenditure that is required to fill the gap within the quality of two assets. Sometimes,
consulting with an architect could serve the purpose. Here, our estimate is that if our
property must reach the same condition as the comparable property, we may have to invest
around $5 million. The subject property’s price (with respect to the first comparable)
should be adjusted down by $5 million. You see that the adjustment of $5 million in the
table is listed as negative.
Further, the subject property has three outdoor pools, but the comparable property
does not have any. After doing some market analysis, the valuer estimates that the value of
three such outdoor pools in this locality is nearly $800,000. Clearly, there is a lot of
subjectivity in coming up with these estimates.
Similarly, across all the criteria that are priced in hotel transactions, the valuer
develops the adjustment estimates. Finally, the adjustments are summed up with their
appropriate signs per comparable asset. The sum of the adjustments in the Comparable One
with respect to our property is -$5.4 million. However, there is a slight difference of scale
between our subject property and the comparable property. Hence, after standardizing the
adjustment (i.e., dividing the -$5.4 million by 250 keys), the net adjustment is -$21,600.
The adjusted price per room for the subject hotel based on the study of Comparable One
would be $292,000 less $21,600 (=$270,400). Repeat this exercise with other comparable
assets. Based on Comparable Two, the adjusted price per room for our hotel should be
around $270,600. If the adjustments are fairly accurate, then we should expect adjusted
prices per key derived from different comparable sales to converge.
Based on some other criteria that cannot be directly observed or quantified, one
could argue that a comparable is more like the subject property than another. Accordingly,
a scheme of “similarity weight” could be assigned to the adjusted prices per key. One must
make sure that the sum of the similarity weights must be 100%. The weighted average sale
price per room for our hotel is $270,760. We multiply this number by the number of keys
in our hotel and that leads us to a value of $81,228,000.
No method for valuing a hotel is perfect. A good sales-comparison approach evaluation
will require very similar comparable sales. For example, if we are valuing a luxury hotel, our
comparable sales should preferably come from the luxury hotel segment. The comparable sales
should also be more or less from the same or very similar locality. It must be of the same scale,
for example, if our subject asset is a 50 sqm office, we cannot possibly compare it with another
office tower of 5000 sqm, although it might fall in the same locality or the same asset class.
Besides, the sale date is extremely important. Real estate valuation is specific to a date. Valuation

Page 45 of 93
of an asset from the past year may not be valid today. If a comparable sale is too old, then a valuer
must use a valuation index to inflate the value along the index and bring the price to the present
levels. Adjusting the prices based on quality differentials requires strong scrutiny.
When selecting comparable sales, one must ensure some conditions, some of which are listed
below:
1. Are the comparable assets really comparable? In other words, are they from the same
market, of similar size, quality, and class?
2. Did the comparable sale transactions occur within the same time frame? The comparable
hotels must not belong to a different economic cycle.
3. Are there any comparable sales that do not reflect a fair market or arm’s length
transactions?
4. If the above three conditions are not perfectly met, are the adjustments made to address the
differences based on acceptable calculations?
Reconciliation of Value
By now, you must be convinced that valuation is art supported by scientific methods. There is
ample room for subjectivity in a valuation exercise. As real estate markets are known to be
inefficient, it is a practical challenge to collect accurate data and build a high degree of confidence
in value judgment. The valuer must make their peace with this reality and make sure to back their
assumptions with documented evidence. It is not rare for valuers to be sued in a court of law by
their clients.
That is one of the reasons why a valuer tries to value an asset using different methods.
Despite the income approach being the most widely accepted method of valuation, it is an industry
practice to adopt some other methods as well. The cost approach and the sales comparison
approach are among the most popular alternatives.
As we have seen earlier, the income approach, the cost approach, and the sales comparison
approach are based on three different philosophies. It is, then, natural to expect that the same asset
may be perceived to have three different values from different viewpoints. As an analyst gains
more experience, the different value estimates start to fall within the same ballpark. Also, when
the market is relatively more efficient in that there are more data available, different methods may
still lead to similar value estimates.
If the value estimates from the three methods turn out to be drastically different from each
other, it is advisable that we revisit the analysis and conduct a sanity check on the calculations and
data. It may be tedious and potentially confusing.
In the end, it is a common practice to finalize the value estimate based on a weighted
average of value estimates derived from different methods. If the values are fairly similar, this
exercise is less critical. If they are not, it is a common practice to assign more weight to the income
approach valuation and the least to the cost approach. Sometimes, it may be OK to assign zero
weight to other methods and focus purely on the income approach.
Even when one has access to unbiased industry reports, unbiased new information, or
unbiased experts, it is good to ask around and seek opinions and feedback on the appraisal analysis
before finalizing it.

Page 46 of 93
Emerging Methods of Valuation

In developed real estate markets where the availability of data is less of a concern, new,
technology-based valuation methods are emerging. These methods use a large set of real estate
transactions data to extract pricing patterns. Some municipalities have applied hedonic-pricing
methods for mass appraisal of homes to decide on an appropriate property tax. In a hedonic
valuation method, analysts apply econometrically38 based regression models and show the real
estate price (or often the logarithm of price) as a linear combination of its measurable (or
observable) characteristics. Recently, hedonic pricing methods have also been applied to CRE
valuation, as more CRE transactions data is available. Beyond econometrics models, some analysts
have proposed machine learning models as Automatic Valuation Methods (AVM). Such methods
of valuation are still emerging and have the potential to be the norm of valuation in the coming
years.
38
Statistics applied to economics is termed as “econometrics.”

Page 47 of 93
Machines vs Humans: Real Estate Valuation During the Coronavirus Crisis
Real estate assets are traditionally valued manually, wherein computer tools are used in a
supportive role. However, more sophisticated, machine learning-based valuations have gained
traction in recent years. As the coronavirus crisis has posed serious challenges to the valuation
community, in this article, I assess the role of machines in asset valuation during such crises.
Machines have learned how to wade through gargantuan data: Structured or unstructured,
visual or textual. Machines are incredibly fast, more accurate, and less biased. In short, machines
can indefatigably take up complex, routine tasks with phenomenal efficiency. Human
intelligence, on the other hand, is prone to biases, fatigue, and cognitive limitations. Although
unparalleled in processing abstraction, humans get bored of routine tasks.
Valuation of Assets
Commercial assets are valued for the future cash flows they promise. Future projections of cash
flows, returns, and risks are slaves to what was observed in the past. Mass produced financial
assets, such as stocks, bonds, or derivatives are identical within a batch, like toothpastes
produced in a factory. Besides, in sophisticated markets, they are “liquid”, implying frequent
buying and selling.
A lot of trading in these assets is characterized by the human perception that dictates the
pricing. As trades happen, pricing patterns emerge. The patterns can be documented using
relatively simple and comprehensible mathematics.
But other less liquid investments, like commercial real estate (or private equity, artwork,
wines, etc.) must be valued individually. Each asset is unique, with few common features within
a "class" (e.g. luxury hotels or Bordeaux wines of 2011 vintage).
As a result, valuation is a burgeoning profession. Based on this theme, I will discuss real
estate valuation in this article.
Traditional Valuation Methods
Valuers use several methods, but let us focus on two prominent ones: The income approach and
the comparison approach.
The income approach encompasses detailed cash flow forecasts and an in-depth assessment
of investment risk in a subject asset. This method requires a great deal of subjective judgment
and human ability to wade through data (or its absence thereof), which may be cognitively
challenging. As a result, the method is often criticized for its inconsistencies.

Page 48 of 93
The comparison approach is less fundamental. It seeks to extract a pattern in which investors
have priced assets with differing qualities. The pricing pattern is then applied to the subject asset
to estimate the likely price a typical investor would pay. The price estimate is posited as the
value. But the comparison approach is a double-edged sword: As each asset is unique, other
assets from which the pricing patterns are extracted may not be truly comparable. This warrants
a large number of "adjustments" to pricing, across all the significantly different attributes of the
assets. Such transactions are scarce and collecting data on a large number of comparable assets
is challenging.
With proprietary databases, the issue of data scarcity, fortunately, has been minimized. But
dealing with the swathe of data becomes a cognitive challenge for human appraisers. This is
where machines come to the rescue.
Valuation by Machine-learning
Computer algorithms are applied to transactions data replete with detailed information on price,
asset characteristics, and geographic attributes. These algorithms could be based on econometric
models or on relatively "black box" data science advancements, such as the artificial neural
network.
When two assets differ in pricing, the difference may be allocated to a large number of
factors across which the two assets are themselves different from each other. In theory, there
could be thousands of these pricing factors. Practically, the factors considered may be less in
number. But even, these "parsimonious" models are difficult to analyze by humans. Computers
are adept at such analysis if sufficient data is available.
First, machines "learn" from the training data of asset transactions. The central idea is to
identify significant patterns in pricing: How do specific factors (e.g. asset characteristics) add to
(or diminish) value? They identify which factors are significant in pricing or which factors do
not matter. More importantly, they are also able to quantitatively "allocate" price differentials
to individual factors (or their complex combinations). In short, with incredible speed and
accuracy, the machine learning approach disentangles asset quality differences into a replicable
combination of pricing factors. Then, the combination is vetted against a "test" data set and the
combination with the least pricing error is the winner. Identifying the "winner" algorithm is the
genius of this valuation method, which becomes the "magic" formula for estimating the price of
an asset that is, otherwise, unknown.
Markets are consensus-seeking voting machines
Valuations based on machine learning are, perhaps, a reality of the future and the current efforts
to operationalize it are, definitely, praiseworthy. At their core, such valuations are an expedition
to seek the past market consensus on valuation. The consensus thus documented is, essentially,
retrospective; a powerful documentation of how the market has priced specific attributes of the
assets.

Page 49 of 93
The machine learning-based valuation method works so long as the market perceptions of
value remain broadly stable. But, what if the market perceptions change quickly? Take
coronavirus, for instance…
Can machines help us in dealing with coronavirus-like predicaments?
Assume hypothetically that the machine detects a 20% price premium assigned by the market
to hotels that have spa facilities. In stable economic scenarios, it may be fair to assign a similar
price premium to the next hotel being valued which also has a spa. But who in the Covid-19 era
prefers a hotel because it has a spa? If the pandemic retains its place in investor memory for
coming times, the 20% perception may not be valid anymore. It may reduce to a lower level, if
not transformed into a discount. To estimate the change, the machine needs to be re-trained. The
new consensus needs to be documented afresh.
Hence, the new training of machines will need new data on latest transactions. These
transactions will be rare and their pricing patterns relatively uninformed. The generalizability of
the current pricing patterns will be questionable, as they may be based on numerous "outlier"
prices. Given their retrospective approach, machines-learning-based valuations, at least as they
stand now, fail to rescue us from difficult predicaments.
Humans versus machines
The recent past does not reflect the extraordinary current period, therefore the machine-learning-
based valuation is doomed to fail in today's times. What we need now is a prospective (forward-
looking) valuation method. We already have one: The income approach. Indeed, the income
approach itself is greatly based on our observations of the past. But the human touch may
potentially promise a way out, especially in terms of estimating the risk metrics. This time, the
assessment of risk must weigh more heavily on the emerging consensus of investors, rather than
the appraisal community itself.
We cannot be sure yet whether the impact of coronavirus on real estate will be long-lasting
or just a blip in the history of asset prices. If it is a short-lived situation, future machine learning
endeavors can comfortably exclude the transactions of current times from its training data or
strongly "control for it". This is how the few years of the recent global financial crisis (2007-09)
have been treated.
Since it has been generally speculated that Covid-19 may fundamentally shift our worldview
on work, travel, living and social interactions, the machines will have to wait longer for human
agents to generate new data and settle the new pricing patterns. Either way, human valuers will
come to rescue the machine's endeavors. For now, we are still in command.
Author: Prashant Das, PhD. (This piece is adapted from an original article published in April
2020 with permission from EHL Insights, Switzerland)

Page 50 of 93
Some recent findings on real estate valuation

Despite emerging evidence of behavioral biases in asset prices, unfortunately, the Income
Approach to valuation is limited in incorporating these biases into the valuation. The DCF based
valuation somewhat assumes a rational market, which is not the case. Besides, inefficiency in real
estate markets may lead to a substantial deviation between the appraised value and transaction
prices. With advancements in econometrically or machine learning (ML) based valuation
techniques, the behavioral biases and other less-known determinants of value could be
incorporated in the value estimates.
Does “Being Different” Say Anything About CRE Valuation?
More so than other commercial real estate asset types, hotels can reap the benefits of market
differentiation both by superior operations and by the characteristics of their property.
Much is known about how to beat the competition through superior operations. Operational
managers might dynamically change their differentiation strategy based on changing market
conditions. However, it is the real estate component that tends to be an overwhelmingly large
part of a hotel’s value. Yet, our knowledge about the pricing impact of real estate differentiation
is very limited.
Number of floors and size of a hotel are prominent architectural features. For investors in
existing assets, a property’s age could also be a differentiator. However, owners cannot easily
change these features, although we show that they significantly affect the property price. In our
research—conducted on a sample of nearly 5,000 hotel transactions in the U.S.—we explored
how the superlative building status, such as being the tallest, the largest, or the oldest, raises or
diminishes value, and whether this pattern varies across contexts.
Analytics-based valuation
Recent advancements in commercial real estate research affirm the utility of a statistical
valuation method called “hedonic pricing.” A sample of hotel transactions is fed into computer
models which incorporate market conditions, decipher the pricing pattern, and disaggregate the
asset price into different attributes of a hotel, such as physical and locational. The pricing model
also suggests if these disaggregated price components are significant. This article reports salient
findings which could enhance our traditional valuation framework. In particular, our study
focuses on whether the superlative status has additional price implications.
Does status matter?
The brand: Beyond the impact of a brand’s underlying characteristics — such as class,
amenities, and location type—and contrary to popular belief, we find few cases in which the

Page 51 of 93
brand name has a significant impact on property price. So, we examined additional proxies for
the status.
Location, media coverage, and the buyer: The status of the location matters, although the
local population often generates insignificant revenue for a hotel. Within a metro area, a CBD
locality may add 40% to 85% to the price compared to a non-CBD hotel of similar
characteristics. Even within a submarket of a metro, a hotel in an “affluent” locality—defined
by ESRI as dominated by Caucasian, high-earning, and middle-aged married couples—
generates 11% to 15% additional value. Increased mention in the media, including newspapers,
TV, and magazines, due to events hosted by an upper-class hotel, is also associated with
significant property price enhancement. These factors, clubbed with the characteristics of a hotel
and market conditions, explain nearly 80% variation in hotel property prices.
Further, more active buyers—those who buy or sell hotels more frequently—tend to
significantly overpay (by 0.2% to 0.3%) for hotels. Investors—usually institutional—who will
not operate a hotel themselves overpay significantly (by nearly 11%) when buying it.
The high- and low-quality segments
The findings presented above offer us rather the “baseline” pricing mechanism. Despite the high
accuracy of hedonic pricing models, the actual transaction prices differ from computer
predictions.
Hotels that sold at prices higher than predicted by the pricing model were placed in what we
call the “premium” segment. So, the hotels which sold for prices less than predicted are in the
“low-quality” segment. Note that these two segments are almost equally applicable across all
hotel classes—luxury, mid-scale, economy, and so on. The attached image depicts the
geographic distribution of premium and low-quality hotels.
It is clear that the premium hotels are scattered all over the map, implying that overpricing
or underpricing is pervasive rather than being a metro- or regional-level phenomenon. So, what
factors characterize the premium and low-quality segments?
Standing out by superlative features
Evidently, traditional pricing models already include a building’s height, size and age, among
other hotel attributes. For example, we know how much extra price comes with an additional
room—all else being the same. But we do not know whether enjoying “the largest” status—for
example, the highest number of rooms—has significant price implications. Similarly, we do not
know the price implications of other superlative statuses, such as the “tallest” or the “oldest.”
Besides, there are discrete scales of superlative comparisons. For example, a hotel might
either be among the tallest nationally or only the tallest in its local market. We define nationally

Page 52 of 93
superlative hotels as those which lie in the top 1% in terms of size, height, or age. Locally
superlative status is identified by comparing the subject hotel with surrounding hotels in its local
trade area—for example, within a 20-minutes driving distance.
We found that the locally largest status is considered an “atypical” feature. Because of their
specificity, there is a thin demand for such hotels. Hence, this status could diminish the price by
up to 15%. On the other hand, nationally largest hotels enjoy distinctive recognition and
capacity, leading up to a 30% price premium. Also, if a locally largest hotel conspicuously stands
out in its size within its locality, the negative effect of this status might become insignificant.
The status of being locally or nationally the tallest has a different impact on pricing. Premium
hotels earn some distinction from being the tallest. Investors in this segment could use this
feature as subtle marketing or “ego-enhancing” tool, as some earlier studies have reported.
However, investors in low-quality hotels see the “tallest” status as a limitation. Low-quality
hotels that are “too tall” offer challenges in fire safety as well as vertical transportation and
create psychological concerns.
A similar pricing pattern can be observed in hotels that are among the oldest nationally. Such
hotels are usually 100 years or older. Premium-segment investors value the historic nature and
assign additional valuation to such hotels. However, investors in the discount segment might be
concerned about the ongoing costs involved with very old assets.
Externality of superlative hotels
We detect another intriguing phenomenon: being locally the oldest, largest, or tallest also affects
the valuation of neighboring assets. Unless its vintage effect conspicuously stands out, the oldest
hotel negatively affects the value in neighboring low-quality hotels. The well-recognized
identity of the oldest hotel in a neighborhood could be seen as a threat to the demand for
neighboring hotels. However, high-quality hotels sometimes enjoy value enhancement if
another hotel in the neighborhood is locally the oldest.
Locally, the largest hotel does not significantly affect the valuation of surrounding lower-
quality hotels. However, if the size conspicuously stands out, higher-quality hotels in the
neighborhood might enjoy higher pricing due to another hotel that is locally the largest. This
could be due to the economy of agglomeration in some markets: Higher-quality hotels are able
to capture demand spillovers from another large hotel in the locality, which potentially attracts
events on a regular basis.
Similarly, the tallest hotel in the locality often does not significantly affect the pricing in
surrounding low-quality hotels. On the other hand, if the height is moderate, the rental demand
from the tallest hotel might spill over to the surrounding premium hotels, which gain some value
from this positive externality. However, if the tallest hotel visibly stands out, this creates a
“snobbery” effect. Such a conspicuously tall hotel could be seen as overpowering the status of
surrounding premium properties, whose prices are, in turn, negatively affected.

Page 53 of 93
Bottom line: One man’s meat is another man’s poison
The impact of these superlative attributes is often economically significant but depends on the
context.
A large hotel must be really large to avoid the “atypicality” pricing discounts or to enjoy the
pricing premium of the “largest” status. Being the tallest or the oldest enhances valuation in the
premium hotels, but often diminishes value in the lower-quality hotels.
Besides, the oldest, largest, or tallest hotels might create both positive and negative externalities
in surrounding hotels depending on the characteristics of the subject hotel. These differences
might stem from economic fundamentals and, sometimes, by psychological biases of investors.
While we now recognize the price implications of superlative real estate attributes, more
research is warranted to improve the accuracy of such models.
Note: Reprinted with Permission from HotelNewsNow. Originally published on 19 July 2017.
This article is based on the following research paper: Das, P., Smith, P. & Gallimore, P.
(Forthcoming). “Pricing Extreme Attributes in Commercial Real Estate: The Case of Hotel
Transactions,” Journal of Real Estate Finance & Economics 151–160. DOI: 10.1007/s11146-
017-9621-4
Summary of the Chapter

"Value" is a philosophical construct that is difficult to define. Investment value (IV) is the price at
which an investor is willing to pay for an asset-based on her perceived utility from the investment.
The market value (MV) of an asset is the price at which it is offered for sale. Buying and selling
decisions should be based on a comparison of MV and IV. The valuation process entails defining
the problem, gathering relevant information, and analysing it. Following this, a valuer reconciles
and includes valuations from alternative scenarios. Finally, all of the data is compiled into a report.
When it comes to valuation methods, there are three to consider: the sales comparison approach,
the cost approach, and the income approach. The income approach calculates the fair value based
on the income generated by the property. Direct capitalization and discounted cash flows are two
methods that help us determine the fair value of an asset using the income approach. The cost
approach assumes that a prospective buyer of a property should pay the cost of constructing an
equivalent building. When a property is new, it produces the most accurate market value than other
methods. Sales Comparison is a more practical approach that focuses on predicting the price at
which an asset with certain characteristics will sell in a market. This method is more commonly
used in residential valuation. It makes no guarantees about estimating an asset's "intrinsic" or
"natural" value. Emerging valuation methods extract pricing patterns from a large set of real estate
transaction data. As more CRE transaction data becomes available, hedonic pricing methods have
been applied to CRE valuation. Machine learning models, in addition to econometric models, have
been proposed by some analysts.

Page 54 of 93
Examine your understanding with the following questions:
1. In which of the following scenarios should the going-out cap rate be lesser than the going-
in cap rate in a 5-year discounted cash flow (DCF) analysis?
a) Today`s hotel investment market is in equilibrium.
b) Today`s hotel investment market is in a bubble.
c) Today`s hotel investment market is in a crisis.
2. Which of the following statements is false? For a given series of stabilized cash flows, the
lower the price you pay for a hotel...
a) …the higher the total return.
b) …the higher the going-in cap rate.
c) …the higher the going-out cap rate.
d) …the higher the appreciation return.
3. Your required return on investment, or in other words, your expected total return is 8% per
year. You conduct a hotel investment analysis and find that the IRR is 8.2%. However, the
net present value of the future cash flows using the discounted cash flow (DCF method) is
negative. Which of the following statements is true?
a) You choose to invest because the IRR is larger than your required return.
b) You choose not to invest because the net present value is negative.
c) There must be an error in one of both calculations.
4. Briefly explain the three primary approaches to real estate valuation.

5. List the issues with DCF Valuation.
6. Compare and contrast the following:
a. Going-in vs Going-Out rate
b. Appraisal Value vs Transaction Price
c. Market Value vs Investment Value.
7. What is Value? Why do assets need to be valued?
Time to put the formulae to work (Numerical):

1. Hotel capitalization rates in New Delhi and Mumbai have been quite stable and are
expected to remain so in upcoming years.
In the luxury segment, Delhi and Mumbai offer 4% and 5% IRR respectively.
The respective capitalization rates are 3% and 4% respectively. Which market is
experiencing a faster growth in NOI?
a) New Delhi
b) Mumbai

Page 55 of 93
c) No difference
d) Insufficient Information
2. Your company paid $115 mi to acquire a hotel this year. The cash flows (NOI) will stabilize
after the end of the fifth year. The going-out cap rate and the market discount rates are
estimated to be 5.8% and 8% respectively. What is the market value of this hotel?
Yea Year 1 Year 2 Year 3 Year 4 Year 5 Year 6
r0
NO $6,000,00 $6,500,00 $6,000,00 $6,500,00 $6,750,00 $7,000,00
I 0 0 0 0 0 0
a) $4,664,255
b) $100,000,000
c) $107,402,233
d) -$4,664,255
3. At the end of 2020, hotel Alpha located in Ahmedabad is available for sale for a price of
Rs15 million. In 2020, the net operating income (NOI) of the hotel was Rs. 0.8 million. In
2021, the NOI of the hotel is expected to be Rs. 0.85 million. What is the going-in cap rate
of the hotel if you buy it today?
a. 5.33%
b. 5.50%
c. 5.67%
d. 5.75%
4. Hotel Palmbeach Inc., is choosing to invest in one of the following office space investment
opportunities. All four office spaces are expected to have the same risk and the same
expected total return of 7.5% per year. Which of the hotels has the highest expected growth
rate?
a) Hotel A, which has a cap rate of 6%
b) Hotel B, which has a cap rate of 7%.
c) Hotel C, which has a cap rate of 8%.
d) Hotel D, which has a cap rate of 9%.
5. You are considering to invest in a hotel which is expected to generate a NOI (before capex)
of $1.0 million over the next 12 months. Capital expenditures over the next 12 months are
expected to be $0.1 million. According to market reports, NOI (before capex)-based cap

Page 56 of 93
rates for comparable hotels are currently 7% per year. What is the market value of the hotel
today using the direct capitalization (or direct cap) approach?
a) $12.86 million
b) $13.88 million.
c) $14.29 million
d) Impossible to answer.
6. At the beginning of 2020 you bought a hotel for Rs.15 million. In 2019, the net operating
income (NOI) of the hotel was $0.8 million. In 2018, the NOI of the hotel is $0.85 million.
What is the going-out cap rate if you sell the hotel at the end of 2018?
a) 5.33%
b) 5.50%
c) 5.67%
d) Impossible to calculate
7. In the Chennai CRE market, hotels allocate nearly 20% of EBITDA towards capital
reserves. The market benchmark capitalization rate is 6%. A stabilized hotel is expected to
generate an NOI (after CapEx) worth $12 mi. What is the value of this hotel?
a) $280 mi
b) $225 mi
c) $335 mi
d) $250 mi
8. Akshara works at an asset management firm. She has been given a task to estimate a hotel
Nofrin's value. The property is projected to have Rs. 20 mi in total revenues during the
following year. Total expenses are projected to be Rs. 14 mi. 3% of the total revenue is
allocated as capital reserves. In the next four year, the annual growth rates in total revenues
and total expenses will be {3%, 2%, 4%, 3%} and {5%, 6%, 4%, 2%} respectively. In the
following years, revenues and total expenses will stabilize at a constant rate of 2.5%.
Discount rate on such investments is estimated at 7.5%. Going out cap rate is 6%. Cost of
sales is usually 3%. What is the estimated hotel value today if investment horizon is 5
years?
9. A property costs Rs. 440,000 and this price had been verified to be an accurate estimate of
the property value. Comparable properties in the neighbourhood have been recently traded
at a 9.5% cap rate. What is the NOI in year 2 if growth rate is 8.7%?
10. The market in Jaipur expects an IRR of 13% from hotels. Estimated growth in the overall
hotel sector's NOI is 3%. A typical hotel is projected to earn Rs.14,500,000 in NOI next
year. What is the estimated value of the hotel based on your theoretically derived
capitalization rate?

Page 57 of 93
11. A real estate asset with projected stabilized cash flows of Rs. 52 mi is selling for Rs. 611.8
mi based on the market cap rate perception. Currently, the discount rate for such projects
is nearly 13%. Your estimated growth in income for such assets is 2% per annum.
a. Comparing the cap rates, is it a good buy? Why?

b. What should be the asset value from your own estimates?
12. Anant has to value Hotel Green Park in Mount Abu. with a holding period of 5 years. The
first projected NOI is Rs. 35 mi and is expected to grow at a constant rate of 1.5%. Going-
in cap rate (adjusted for capex) is 5.75% and cap rate spread is 0.50%. Assume there is no
transaction costs and cash flows are stabilized.
a) What is the expected market value according to the direct-capitalization method?
b) What is the expected market value today without considering the holding period
issue? Use the DCF approach.
References
Cook, S. (2018). Investing in Real Estate Private Equity: An Insider’s Guide to Real Estate
Partnerships, Funds, Joint Ventures & Crowdfunding. Kindle Independent.
Das, P. (2015). Revisiting the hotel capitalization rate. International Journal of Hospitality
Management, 46, 151-160.
Das, P., Freybote, J., & Blal, I. (2020). The Importance of Micro-Location for Pricing Real Estate
Assets: The Case of Hotels. Journal of Real Estate Portfolio Management, 1-17.
Das, P.; Kozorez, E.; and Thakare, R. (2020). Sensitivity of Commercial Real Estate Valuation to
Holding Period, and How to Address This. Real Estate Finance.
Das, P., Smith, P., & Gallimore, P. (2018). Pricing extreme attributes in commercial real estate:
The case of hotel transactions. The Journal of Real Estate Finance and Economics, 57(2), 264-
296.
Geltner, D. (2015). Real estate price indices and price dynamics: an overview from an investments
perspective. Annual Review of Financial Economics, 7, 615-633.
Mishkin, Frederic S., 2016, The Economics of Money, Banking, and Financial Markets, eleventh
edition, New York: Pearson Education.
John W. O’Neill, “Hotel Occupancy: Is the Three-Year Stabilization Assumption
Justified?”Cornell Hospitality Quarterly 52, no. 2 (2011): 176-180.

Page 58 of 93
Pilbeam, Keith, 2010, Finance & Financial Markets, third edition, Basingstoke: Palgrave
Macmillan.
Poretti, C., & Das, P. (2020). What explains differing holding periods across hotel investments? A
hazard rate framework. International Journal of Hospitality Management, 89, 102564.
Appendix
Ex-Ante Growth Rate
In corporate finance, analysts argue that in the long run, a firm’s cash flow will broadly align
themselves with the growth rate in GDP. Indeed, the market forces should moderate sustained
deviation from the GDP growth rate. Firms grow as they have the ability to reinvest some part of
the earnings into assets that promise superior growth. Real estate valuation differs from firm
valuation, as the “plowback39” of capital is not a consideration. All cash flows generated from an
asset must be valued without any part of it being reinvested into assets with different returns.
Therefore, the growth rate estimates follow a different method based on detailed cash flow
projections. Forecasting future cash flows involves a rigorous process of forecasting each line item
in the cash flow item that is required for the NOI estimate. As such, the NOI estimate every year
will be a result of these numerous annual forecasts.
An analyst could calculate the resulting growth rate (𝑔) in NOI forecasts after the fact (i.e.
forecasts). For simplicity, let us call it ex-post (“after the fact”) growth rate (𝑔𝑒𝑥 𝑝𝑜𝑠𝑡 ) forecast40.
For various reasons, it is also imperative to estimate the growth rate before the cash flow estimates
are developed. Let us call it ex-ante (“before the fact”) growth rate (𝑔𝑒𝑥 𝑎𝑛𝑡𝑒 ) forecast. Analysts
have different views on what the 𝑔𝑒𝑥 𝑎𝑛𝑡𝑒 should be. In equity stock valuation, several analysts use
the long-term growth rate in GDP as an estimate for 𝑔𝑒𝑥 𝑎𝑛𝑡𝑒 , a method somewhat applicable to
any sustainable cash flow stream. Any asset with 𝑔𝑒𝑥 𝑎𝑛𝑡𝑒 that is substantially below the GDP
growth rate will not remain profitable and must shut down. Other assets whose 𝑔𝑒𝑥 𝑎𝑛𝑡𝑒 is
substantially high may not sustain it in a very long run, as the arbitrageur actions will bring it back
to normal. We can observe these trends in the US cap rate (for selected markets) plotted below41
(Note that the growth rate equals the spread between the discount rate and cap rate):
Like cap rates, discount rates in all segments spiked up as a response to the 2001 dot-com
crisis and 2007 subprime crisis. Overall, we can observe a downward trend in discount rates since
2001. Such trends may not persist forever, and the rates must stabilize at some level; as in the long
run, we do not expect these rates to fall to negative values. In a market, the spread between Ω and
39
Plowback refers to a firm’s manager retaining some part of the earnings to reinvest in assets with superior returns.
40
Note that it is a bit pretentious to call it “after the fact” as the 𝑔𝑒𝑥 𝑝𝑜𝑠𝑡 can only be observed when actual NOIs are
observed in the future. Yet, just to simplify our narrative, let us consider the cash flow forecasts as our “facts” for the
time-being.
41
Based on data from integra realty resources.

Page 59 of 93
cap rate reflects the market’s perception of the growth rate 𝑔. In theory, the Ω is the WACC of a
typical investor for an asset type. As such, a market survey of WACCs should provide a similar
result. In the dearth of relevant data, some analysts also apply the risk-premium approach to
estimating the Ω for valuation. However, a more appropriate method of estimating Ω would be to
sum the 𝛾𝑖𝑛 (after all adjustments) and the 𝑔𝑒𝑥−𝑎𝑛𝑡𝑒 . By now, we have discovered various methods
to estimate both metrics.
Ex-Ante Growth Rate Estimation

Given the depreciable nature and finite life of the current use of a CRE asset, real estate analysts
take a slightly different view on estimating the growth rate. It is argued that due to functional and
external obsolescence, the operating cash flow (NOI) from older assets will gradually slow down.
Depending on the location, older properties may grow one to three points lower than brand new
assets. Besides, the change in inflationary regimes may also impact the future growth rate. Here is
a framework for growth rate prescribed in some textbooks, which, at best, is a broad
approximation:
𝑔𝑒𝑥−𝑎𝑛𝑡𝑒 = 𝑔ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 + ∆𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 − 𝐴𝑔𝑒𝑖𝑛𝑔 𝐸𝑓𝑓𝑒𝑐𝑡
The historical growth rate (𝑔ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 ) should be calculated from past information on the
NOI of similar properties (or the same property in the market). CAGR of the NOI could be
calculated covering several years of data. ∆𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 is the difference in the future forecast of
inflation (during the holding period) in excess of past inflation. The Ageing effect is mostly applied
anecdotally, but could also be estimated using simple market data, as we soon describe.
A challenge with estimating the 𝑔ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 is that the past data on NOI may be scarce, or
unavailable. In such cases, the CAGR of past (1) EBITDA, (2) revenue, or (3) rental rates could
be used as a proxy for growth in NOI (in an order of priority). The lower one goes on the priority
list, the more error is infused in the growth rate estimation. These approximations may be based
on some strong assumptions that may not be reflective of reality. Here are some of those
assumptions in using different proxies for the ex-ante growth rate (𝑖. 𝑒. 𝐶𝐴𝐺𝑅ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 𝑁𝑂𝐼 ):
Example: Average rental rates in Mumbai offices during the last few years ($/sqm/month)
were observed as follows:
Year 2014 2015 2016 2017 2018 2019

Rental Rate 5.8 5.68 5.56 7.99 8.19 8.8
Average inflation in the last 10 years had been around 7.5%. In the next ten years, inflation
is expected to fall to 6%. The oldest properties in the city experience up to 3 points less
growth in cash flows. What is your estimated ex-ante growth rate for a mid-aged office
building in Mumbai? Assume that NOI is broadly proportional to the rental rates.
Solution:
𝑔ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 can be estimated from the CAGR in rental rates over 5 years between 2014 and
2019:

Page 60 of 93
𝑔ℎ𝑖𝑠𝑡𝑜𝑟𝑖𝑐𝑎𝑙 = 8.7%
∆𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 = 6% − 7.5% = −1.5%
−3%
𝐴𝑔𝑒𝑖𝑛𝑔 𝑒𝑓𝑓𝑒𝑐𝑡 = 2 = −1.5%
Therefore, 𝑔𝑒𝑥−𝑎𝑛𝑡𝑒 = 8.7% − 1.5% − 1.5% = 5.7%
Example: Consider all data in the same example as above except that you want to be more
methodical about the ageing effect. Your subject asset is 15 years old. Currently, 30-year
old hotels in the locality have an average rental rate of $8.6/sqm/month whereas brand new
offices in the same class are charging $8.99/sqm/month.
1
8.6 (30)
Solution: deceleration in rental rates attributed to age is (8.9) − 1 = −0.11%
𝐴𝑔𝑒𝑖𝑛𝑔 𝑒𝑓𝑓𝑒𝑐𝑡 = (1 − 0.11%)15 − 1 = −1.7%
Therefore42, 𝑔𝑒𝑥−𝑎𝑛𝑡𝑒 = 8.7% − 1.5% − 1.7%% = 5.5%
Note that the ageing effect, at best, is a broad estimation without much scientific support.
One must apply contextual information to justify the negative effect of a property’s age on its cash
flow growth, as well as the quantification of this effect.
Growth Rate Dilemma
The discount rate is reflective of the capital market where investors build their expectations of total
return. An investor can settle with a smaller income return (cap rate) if the cash flow growth rate
is higher. Theoretically, the cap rate is a function of growth rate (cap rate = discount rate – growth
rate), something determined in the asset market. For two assets with the same stabilized NOI, one
should expect a higher valuation from the one that has a higher ex-post growth rate. However, a
higher rate implies a higher discount rate which will have an opposite effect on valuation, as the
growth rate is additive to the discount rate when calculated from the going-in cap rate. Such a
derivation of discount rate may be counter-intuitive.
The VastraMart valuation case, in a way, was an idealistic view of cash flows where all
growth rates were the same. In Pro-forma, the NOI projections are an outcome of a large set of
other revenue and expense variables. The CAGR of projected NOI serves as our projections for
the ex-post growth rate. After stabilization, the future NOIs should grow at the ex-post rate. As
cash flow projections themselves may depend on an individual’s subjective perceptions, the ex-
post growth rate projection may deviate from the ex-ante growth rate. This leads to an undesirable
valuation paradox (i.e., a higher ex-ante growth rate leads to a higher discount rate).
Therefore, it is important to make sure that the ex-ante growth rate (that is used for discount
rate calculation) is not far off from the ex-post growth rate (that is derived from NOI projections).
In other words, if the ex-ante growth rate is high (or low), the ex-post growth rate must also be of
42
Note that for modest rates, one could approximate the ageing effects by simple multiplication: 0.11% × 15 ≈
1.7%

Page 61 of 93
the same nature. In practice, this endeavor warrants taking a careful look at the cash flow Pro-
forma and adjusting some revenue or expense items in such a fashion that the two growth rates are
similar.

Page 62 of 93
Figure 5.1 Reservation Price Frequency Distribution of Individual Buyers and Sellers

Page 63 of 93
Figure 5.2

Page 64 of 93
Figure 5.3: Rental Rates in Selected Indian Cities (2020)
RE TAIL RE NTAL RAT E S IN 2020

($/ S Q M/ MO NT H )
Min Max
122
53.33
72
44.67
22.93
19.33
17.33
32
32
24
MUMBAI GURGAON HYDERABAD BANGALORE CHENNAI
DATA SOURCE: BMI

Page 65 of 93
Figure 5.4:
Capitalization Rate in Selected Market Segments

14
13
12
11
10
5
2010 2011 2012 2013 2014 2015 2016
Data source: BMI
Mumbai Office Gurgaon Office Hyderabad Office Bangalore Office Chennai Office
Mumbai Retail Gurgaon Retail Hyderabad Retail Bangalore Retail Chennai Retail

Page 66 of 93
Figure 5.5: Spread of Risk Premium in Selected Office Markets in India (2010-2016)
Mumbai Office Gurgaon Office Hyderabad Office

Bangalore Office Chennai Office
8
6
4
Spread (% points)
2
0
-2 2010 2011 2012 2013 2014 2015 2016
-4
Data source: BMI

Page 67 of 93
Figure 5.6: Risk premium across the globe

Page 68 of 93
Figure 5.7: Valuation Metrics for New York Hotels
Valuation Metrics for New York Hotels

11.90%
10.90%
9.90%
8.90%
7.90%
6.90%
5.90%
Discount Rate (%) Going-In Cap Rate (%) Going-Out Cap Rate (%)
Data Source: Intergral Realty Resources

Page 69 of 93
Figure 5.8: Process of Determining Cap Rate for CRE Valuation

Page 70 of 93
Figure 5.9: Estimating the Cap Rate

Page 71 of 93
Figure 5.10: Scatterplot between Transaction-Based and Appraisal-Based Cap Rates
Appraisal versus Transaction Cap Rate

21.5
19.5 R² = 0.6288
17.5
Transaction Cap Rate
15.5
13.5
11.5
9.5
7.5
5.5
3.5
3.5 8.5 13.5 18.5 23.5
Appraisal Cap Rate

Page 72 of 93
Figure 5.11: Rental Apartment Market in Atlanta, USA (NCREIF Data)
0.1
0.09
0.08
0.07
0.06
0.05
0.04
19964
19973
19982
19991
19994
20004
20013
20022
20031
20034
20043
20052
20061
20064
20073
20082
20091
20094
20103
20112
20121
20124
20133
20142
20151
20154
20163
20172
20181
20184
20193
20202
Appraisal-based Transaction-based

Page 73 of 93
Figure 5.12: Probabilities of different HPs based on a sample of over 100,000 CRE Assets

Page 74 of 93
Figure 5.13: Variation in Holding Periods across property types

Page 75 of 93
Figure 5.14: Variation in Holding Periods within subclasses of an asset class

Page 76 of 93
Figure 5.15: Evolution of Rental Rates in Some Leading CRE Markets in India
CRE Rental Rates in Selected Market Segments ($/Sqm/Month)
120
100
80
60
40
20
0
Data Source: BMI
2014 2015 2016 2017 2018 2019 2020

Page 77 of 93
Figure 5.16: Sensitivity of CRE Valuation to the Assumed Holding Period
Sensitivity of CRE Valuation to the Assumed

Holding Period
$126,000,000 0.00%
Deviation in Appraised Value from

$124,000,000
-2.00%
Directly capitalized valuation

$122,000,000
Appraised value
$120,000,000
$118,000,000 -4.00%
$116,000,000
$114,000,000
-6.00%
$112,000,000 -8.00%
$110,000,000
$108,000,000 -10.00%
$106,000,000
$104,000,000 -12.00%
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Assumed Holding Period (HP)
Other assumptions | y.in=10%, y.out= 11.5%, g = 5%, NOI.year1 = $1.25

mi

Page 78 of 93
Figure 5.17: Sensitivity of appraised value

Page 79 of 93
Figure 5.18: Monte Carlo Simulation of the Appraised value
Monte Carlo Simulation of Appraised Value

450
400
350
FREQUENCY
300
250
200
150
100
50
0
11.0
11.4
11.8
12.2
12.6
13.0
13.4
13.8
14.2
14.6
15.0
15.4
15.8
16.2
16.6
17.0
17.4
17.8
18.2
18.6
19.0
19.4
19.8
VALUE (₹ CRORE)

Page 80 of 93
Figure 5.19: Geographic distribution of premium and low-quality hotels. Warm colors—usually
in the center of clusters—such as yellow, orange and red suggest higher concentration of premium
hotels. (Illustration: Prashant Das and ESRI)

Page 81 of 93
Figure 5.20: Trends in the US Cap Rate

Page 82 of 93
Table 5.1:
𝑃𝐺𝐼 Potential Gross Income ₹19,200,000
−𝑉𝐴𝐶 Vacancy Loss - ₹192,000
+𝑀𝐼 Miscellaneous Income + ₹3,000,000
= 𝐸𝐺𝐼 Effective Gross Income = ₹20,280,000
−𝑂𝑃𝐸𝑋 Operating Expenditure - ₹5,070,000
= 𝐸𝐵𝐼𝑇𝐷𝐴 Asset-specific Earnings Before Interest, Tax, = ₹15,210,000
Depreciation, and Amortization
− 𝐶𝑎𝑝𝑅𝑒𝑠 Capital Reserves - ₹1,014,000
= 𝑁𝑂𝐼 Net Operating Income = ₹14,196,000

Page 83 of 93
Table 5.2: Composition of Cap Rate

Risk-Free Rate 4.5%
Sources of Risk (Uncertainty) Risk Premium
1. Occupancy 1.0%
2. Quality 1.0%
3. Ageing 0.5%
4. Local Demographics 0.5%
5. Marketability 1.0%
6. Tenant Quality 1.0%
7. Tenant Mix 1.0%
Cap Rate 10.5%

Page 84 of 93
Table 5.3: Cash Flow Proforma

Page 85 of 93
Table 5.4: Cash Flow Proforma

Page 86 of 93
Table 5.5:

Page 87 of 93
Table 5.6: Monte Carlo Analysis

Ω 𝑪𝑭𝟐𝟎𝟏𝟖 𝑪𝑭𝟐𝟎𝟏𝟗 𝑪𝑭𝟐𝟎𝟐𝟎 𝑪𝑭𝟐𝟎𝟐𝟏 𝑪𝑭𝟐𝟎𝟐𝟐
Mean 13.0% ₹14,763,840 ₹15,354,394 ₹15,968,569 ₹16,607,312 ₹181,499,891
Min 10.0% ₹12,000,000 ₹12,500,000 ₹13,000,000 ₹13,500,000 ₹150,000,000
Max 16.0% ₹16,000,000 ₹17,000,000 ₹18,500,000 ₹20,000,000 ₹210,000,000
Stdev 1.5% ₹1,000,000 ₹1,125,000 ₹1,375,000 ₹1,625,000 ₹15,000,000

Page 88 of 93
Table 5.7:
SN Discount Rate CF(year1) CF(year2) CF(year3) CF(year4) CF(year5) Value
1 13.0% ₹14,763,840 ₹15,354,394 ₹15,968,569 ₹16,607,312 ₹181,499,891 ₹144,853,552
2 12.5% ₹12,436,596 ₹14,745,865 ₹16,103,997 ₹14,108,063 ₹234,143,099 ₹172,926,304
3 11.5% ₹14,634,324 ₹16,880,661 ₹14,168,572 ₹16,646,767 ₹155,125,003 ₹137,919,011
4 13.9% ₹15,196,278 ₹15,980,251 ₹17,227,067 ₹15,687,988 ₹188,838,982 ₹145,339,644
5 11.3% ₹14,606,588 ₹13,775,540 ₹15,044,776 ₹18,461,031 ₹189,336,272 ₹157,994,838
⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮
4997 13.7% ₹14,554,316 ₹15,083,501 ₹15,857,339 ₹16,905,942 ₹185,049,077 ₹142,610,401
4998 10.7% ₹15,209,535 ₹15,919,622 ₹15,786,718 ₹14,538,020 ₹205,469,128 ₹171,924,114
4999 13.8% ₹14,680,996 ₹14,519,692 ₹12,859,199 ₹16,251,156 ₹159,031,495 ₹125,806,330
5000 13.0% ₹13,736,473 ₹16,067,467 ₹16,559,874 ₹16,822,782 ₹201,024,334 ₹155,907,430

Page 89 of 93
Table 5.8:
Property type Units Property Type Units
Hotel Room (Key) Tennis Facility Court
Restaurant Seat Hospital Bed
Golf Course Hole Warehouse Truck door
Factory Gross Building Area Marina Slip43
43
Parking space for boats

Page 90 of 93
Table 5.9:
Gross Revenue Price Gross Revenue Multiplier (GRM)
Office A ₹ 120,000 ₹ 1,500,000 12.50
Office B ₹ 370,000 ₹ 5,000,000 13.51
Office C ₹ 230,000 ₹ 3,540,000 15.39
Office D ₹ 540,000 ₹ 6,134,000 11.36
Average GRM 13.19

Page 91 of 93
Table 5.10:

Page 92 of 93
Table 5.11: Assumptions
Proxy for
𝒈𝒆𝒙−𝒂𝒏𝒕𝒆 Assumption Comment
𝑪𝑨𝑮𝑹𝑬𝑩𝑰𝑻𝑫𝑨 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝑟𝑒𝑠𝑒𝑟𝑣𝑒𝑠
A. 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 Maybe broadly true in many cases
𝐸𝐵𝐼𝑇𝐷𝐴
𝑪𝑨𝑮𝑹𝑹𝒆𝒗𝒆𝒏𝒖𝒆 Assumption A, and Maybe somewhat true in normal
𝐸𝑥𝑝𝑒𝑛𝑠𝑒𝑠 market conditions when fixed expenses
B. 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
are in a small proportion
𝑪𝑨𝑮𝑹𝑹𝒆𝒏𝒕𝒂𝒍 𝑹𝒂𝒕𝒆 Assumptions A, B, and May not be true in most cases as
C. Occupancy rate is stable Occupancy and rental rates follow
different market cycles.

Page 93 of 93
Table 5.12: Example of Cash Flows in New York Luxury Hotels

Reading-Valuaton Methods

Uploaded by

Copyright:

Available Formats

Reading-Valuaton Methods

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reading-Valuaton Methods

Uploaded by

Copyright:

Available Formats

Page 1 of 93

At the end of this chapter, the reader will be able to understand:

Page 1 of 93 ©Prashant Das 2021. Do not distribute.

Page 2 of 93 ©Prashant Das 2021. Do not distribute.

Need for Valuation

Page 3 of 93 ©Prashant Das 2021. Do not distribute.

Valuers & Regulation

CRE Valuation as a Profession

Page 4 of 93 ©Prashant Das 2021. Do not distribute.

A Historical Background of Valuation Professionals

Regulation of Valuation Profession in India

Page 5 of 93 ©Prashant Das 2021. Do not distribute.

Page 6 of 93 ©Prashant Das 2021. Do not distribute.

5. Scenarios: A good valuation analysis includes analysis of alternative scenarios and

6. Report Development: Finally, all relevant information should be consolidated in a well-

Page 7 of 93 ©Prashant Das 2021. Do not distribute.

Refer to the exhibit 5.2 below.

Page 8 of 93 ©Prashant Das 2021. Do not distribute.

Mathematically, the Value of the asset at present (𝑛 = 0) equals

Page 9 of 93 ©Prashant Das 2021. Do not distribute.

Estimating the EBITDA

Page 10 of 93 ©Prashant Das 2021. Do not distribute.

Estimating the NOI

More on Capital Reserves

Page 11 of 93 ©Prashant Das 2021. Do not distribute.

Valuation Using the Direct Capitalization Method

Page 12 of 93 ©Prashant Das 2021. Do not distribute.

Page 13 of 93 ©Prashant Das 2021. Do not distribute.

Nature of Capitalization Rate

Page 14 of 93 ©Prashant Das 2021. Do not distribute.

Page 15 of 93 ©Prashant Das 2021. Do not distribute.

Estimation of Cap Rate

When a Cap Rate Benchmark is Available

Page 16 of 93 ©Prashant Das 2021. Do not distribute.

Page 17 of 93 ©Prashant Das 2021. Do not distribute.

When a Cap Rate Benchmark is Not Available

First Principles Approach

Page 18 of 93 ©Prashant Das 2021. Do not distribute.

20 𝑎𝑛𝑛𝑢𝑎𝑙 𝑚𝑜𝑟𝑡𝑔𝑎𝑔𝑒 𝑟𝑒𝑝𝑎𝑦𝑚𝑒𝑛𝑡 𝑎𝑚𝑜𝑢𝑛𝑡

Page 19 of 93 ©Prashant Das 2021. Do not distribute.

Cap Rate Adjustment for Capital Reserves

Page 20 of 93 ©Prashant Das 2021. Do not distribute.

Going-In versus Going-Out Cap Rate

Page 21 of 93 ©Prashant Das 2021. Do not distribute.

Page 22 of 93 ©Prashant Das 2021. Do not distribute.

1. 𝛾𝑖𝑛 evolves over time due to market conditions.

𝛾𝑜𝑢𝑡 = 9% − 0.5% + 0.25% = 8.75%

Page 23 of 93 ©Prashant Das 2021. Do not distribute.

𝛾𝑜𝑢𝑡 = 9% − 3% + 0.15% = 6.15%

Case Study: “Jane Decides on Cap Rate”

Page 24 of 93 ©Prashant Das 2021. Do not distribute.

Further, we have a piece of subjective information that McDavid’s central locality

Page 25 of 93 ©Prashant Das 2021. Do not distribute.

Going-Out Cap Rate

Appraisal versus Transaction Cap Rates

Page 26 of 93 ©Prashant Das 2021. Do not distribute.

different from an investor’s investment plan. The NOI, growth rate,

Page 27 of 93 ©Prashant Das 2021. Do not distribute.

Discounted Cash Flow Method