Investment Course - 2005

Day Two:

Equity Analysis and Portfolio Strategies

2-0
 Forming Equity Portfolios: An Overview
 After an investor’s strategic asset allocation (i.e., the percentage allocations
to the broad asset classes) has been established, the next step in the
portfolio management process is to form asset class-specific portfolios that
we think will align with our investment objectives and constraints.

 Asset class-level (e.g., stock) portfolios can be formed by following one of

two approaches:
 Passive: Designed to match a broad equity index (e.g., S&P 500, Russell 1000,
IPSA); implicitly assumes that equity markets are efficient
 Active: Attempts to outperform a designated equity benchmark, usually through
picking stocks perceived to have superior characteristics (e.g., valuation, style)

 Generally speaking, active equity management can be approached in one

of two ways:
 Top-Down (i.e., Three-Step) Approach
 Bottom-Up (i.e., Stock-Picking) Approach
The difference between the two approaches is the perceived importance of
macroeconomic and industry influences on individual firms and stocks

2-1
 The Three-Step Valuation Process
1. General economic influences
 Decide how to allocate
investment funds among
countries, and within countries
to bonds, stocks, and cash
2. Industry influences
 Determine which industries
will prosper and which
industries will suffer on a
global basis and within
countries
3. Company analysis
 Determine which companies in
the selected industries will
prosper and which stocks are
undervalued

2-2
 Example of a Global Portfolio:
Texas Teachers Retirement System - September 30, 2004

2-3
 Example of a Global Portfolio (cont.):
Texas Teachers Retirement System - September 30, 2004

2-4
 Examples of Style-Based Equity Portfolios

A. Babson Growth Fund (BABSX)

Market Cap Est Growth Div. Yld.

Company Ticker ($ Bil) P/E P/BV EPS (%) (%) Beta
Fed Home Loan Mtg FRE 45.0 15.39 4.18 14.48 1.36 0.51
Pfizer PFE 249.3 30.37 13.65 19.53 1.31 0.67
Kinder Morgan KMI 6.2 25.35 2.74 19.89 0.40 0.55
Citigroup C 254.1 17.57 3.20 14.39 1.29 1.16
American Intl Group AIG 189.7 28.79 3.64 14.18 0.23 0.79
Exxon Mobil XOM 301.8 19.75 4.14 7.45 2.07 0.61
Medtronic MDT 53.7 37.85 8.72 17.17 0.52 0.88
Microsoft MSFT 314.0 32.39 6.08 15.79 0.00 1.12
Symantec SYMC 5.7 33.04 4.50 20.08 0.00 1.27
Paychex PAYX 14.5 53.78 16.44 20.49 1.14 0.86

Average: 143.4 29.43 6.73 16.35 0.83 0.84

B. T. Rowe Price Value Fund (TRVLX)

Market Cap Est Growth Div. Yld.

Company Ticker ($ Bil) P/E P/BV EPS (%) (%) Beta
Honeywell Intl HON 31.3 18.73 3.41 13.20 1.95 1.27
Bank One ONE 48.7 17.18 2.41 11.30 2.01 0.93
RadioShack RSH 5.1 19.25 7.25 15.21 0.75 1.31
Schering-Plough SGP 43.9 18.95 6.16 11.64 2.14 0.53
American Express AXP 53.6 30.30 4.42 12.62 0.80 1.30
Exxon Mobil XOM 301.8 19.75 4.14 7.45 2.07 0.61
Allstate ALL 26.8 17.83 1.56 10.23 2.23 0.76
Burlington Resources BR 8.4 12.94 2.37 21.30 1.32 0.58
Bristol-Myers Squib BMY 74.3 15.39 6.92 10.36 2.92 0.55
May Dept Stores MAY 10.0 15.50 2.62 9.65 2.76 0.79

Average: 60.4 18.58 4.13 12.30 1.90 0.86

2-5
 General Approaches to Equity Valuation
1. Discounted Cash Flow (DCF) Valuation: A firm is worth the net
present value of the cash flows it is expected to provide to its stakeholders

- Discounted Dividends (D) or Earnings (E)

- Discounted Free Cash Flow to Equity (FCFE)
- Discounted Free Cash Flow to Firm (FCFF)
(i.e., Enterprise Value)
- Adjusted Present Value (APV) and Capitalized Earnings (CE)

2. Relative Valuation Techniques (“Comparables”): A firm’s value is

determined by comparing it to the value of similar companies

- Price/Earnings Ratio (P/E)

- Price/Book Value Ratio (P/BV)
- Price/Free Cash Flow Ratio (P/FCF)
- Price/Sales Ratio (P/S)
- Enterprise Value/EBITDA (EV/EBITDA)

3. Acquisition Valuation: A calculation of what the company in question

would be worth to a third-party acquirer

- Takeover Pricing
- Breakup Analysis
- Comparable Transaction Analysis

4. Technical Valuation: Short-term future share prices established from

prior trading patterns

- Price Charting
- Technical Indicators

5. Option-based Valuation: Applies sophisticated derivative security

valuation techniques to value the cash flows generated by a firm

- Probabilistic Forecasts of Future Events

- Near-Bankruptcy Valuation
- “Real” Option Problems
2-6
Enterprise Value vs. Equity Value

2-7
 The Foundations of Stock Valuation
The general equation for establishing the fundamental value of a financial asset:

N
CF P
P0   (1  k)t t  (1  Nk) N
t 1

where CFt is the period t cash flow while holding the asset and PN is the asset’s period N terminal
value. This equation can be applied to the valuation of fixed-income securities in a relatively
straightforward manner since a bond contract typically specifies both the periodic and terminal
cash flows as well as the payment dates.

There are, however, two problems that prevent a simple application of the formula to the valuation
of common stock:

1. Stock does not mature, meaning that there is no definitive terminal price (i.e., face value) and, at
least theoretically, an infinite stream of future cash flows;

2. Stockholders are residual claimants, meaning that the future cash flows (e.g., dividends) are
neither guaranteed nor promised in advance.

The first problem can be addressed by recognizing that the period N value of the stock (P N) should
be the present value of the cash flows starting in period N+1. Thus, the value of the stock today
can be expressed as:

N
CF  1   CFN  t 
P0   (1  k)t t 
 (1  k) N
 



 t 1 (1  k)
t
t 1  

or:


CF
P0   (1  k)t t .
t 1

That is, the period 0 value of the stock is the present value of all future cash flows; no terminal
price estimate is necessary regardless of a given investor’s target holding period.

2-8
The Foundations of Stock Valuation (cont.)
The second problem involving the specification of an infinite number of unknown cash flows is
obviously more problematic. What many analysts prefer to do is to focus on predicting changes in
the level of the cash flows from one period to the next rather than the levels themselves. That is,
the preceding valuation equation can be altered to focus on cash flow growth rates instead of the
dollar levels of the expected cash flows. This can be done by expressing every future forecasted
cash flow in terms of the current observable level (CF0) and the appropriate sequence of transitional
growth rates (gt):

Period Cash Flow

0 CF0
1 CF1 = CF0(1+g1)
2 CF2 = CF0(1+g1)(1+g2)
. .
. .
. .
N 
N CFN  CF0   (1  g t ) 
 t 1 

so that:

 CF0 t (1  g t ) 
t (1  g t )
P0   (1  k) t
 CF0  (1  k) t
t 1 t 1

Notice that this valuation formula is still infinite-lived, but now it does not depend on the explicit
forecast of future cash flow levels.

There are three common assumptions that analysts use to describe how cash flows change over
time:

1. Constant Growth (all gt = g)

2. No Growth (g = 0)
3. Multi-Stage Growth (all gt = g only after an initial period of “abnormal”
growth)
2-9
The Foundations of Stock Valuation (cont.)
 Constant Growth:

The constant growth formula should be considered the base case for all discounted cash flow
valuation models. With the assumption that gt = g for each future period, the above valuation
model reduces to:

 
CF0 (1  g) t (1  g) t
P0   (1  k) t
 CF0  (1  k) t
t 1 t 1

The advantage of this static growth assumption is that the valuation problem reduces to the
designation of two forecast variables: k (the cost of capital) and g (the constant cash flow growth
rate).

Even with this simplification, though, the above valuation equation still requires the summation of a
countless number of discounted cash flows. This “infinite life” problem can be addressed by
assuming that k > g, which insures that future cash flows will be discounted back to the present
more rapidly than they grow from one period to another. Said differently, assuming k > g
guarantees that at some point the present value of the future cash flows growing at the constant rate
g becomes zero. It can be shown that with this additional assumption, the above constant growth
valuation specification becomes:

CF1 CF0 (1  g)
P0  
k -g k-g

 No Growth:

Notice that the “no growth” version of the valuation model—that is, where g = 0—is just a special
case of the constant growth formula. Importantly, this specification is used to value stocks that pay
a constant, perpetual dividend (e.g., regular preferred stock). In fact, with g = 0 the constant growth
model reduces to:

CF0
P0 
k

which is just the formula for the present value of a perpetuity.

2 - 10
Constant Growth Valuation Example: CSR

2 - 11
Constant Growth Valuation Example: CSR (cont.)

2 - 12
Constant Growth Valuation Example: CSR (cont.)

2 - 13
Constant Growth Valuation Example: CSR (cont.)

2 - 14
The Foundations of Stock Valuation (cont.)
 Multi-Stage Growth:

Few companies generate cash flows that grow at anything close to a constant rate during their entire
life cycles. Nevertheless, the collective assumptions of the constant growth model (i.e., a constant g
that is less than k) are necessary for the valuation problem to “collapse” to a tractable form. One
compromise is to designate at least two distinct periods of growth: an initial stage where growth
rates can vary period by period and a terminal stage where cash flow growth becomes constant and
less than k.

With this modification, the multi-stage valuation model becomes:

N CF0 [  1 + g t  ]  CFN (1 + g 2 ) 
  1 + k N 
-N
P0 = + 
t =1 1 + k t 
t
 k N - g 2  

= PV of Stage 1 (“Abnormal”) Growth

+ PV of Stage 2 (“Constant”) Growth

In this specification, the length of the initial growth stage is N periods and is a variable that must be
specified by the analyst. Further, since the cash flows in the first stage must be discounted and
summed separately, it is possible for each period to have a different growth rate and a different
discount rate. Here g2 and kN represent the constant growth and discount rates in the terminal stage
that begins with Period N+1.

Notice that if in Stage 1 all gt = g1, the multi-stage model reduces to:

N
CF0 1 + g1 t  CFN (1 + g 2 ) 
  1 + k N 
-N
P0 = + 
1 + k t   k N - g 2  
t
t =1

This form of the valuation equation is sometimes called the two-stage growth model because it
allows for two separate constant growth rate regimes: g1 for the first N periods and g2 thereafter.
Finally, notice that in the terminal stage of growth, the discount rate applied to each cash flow (kN)
must be constant as well.
2 - 15
The Foundations of Stock Valuation (cont.)
Although the preceding assumptions reduce the two-stage growth model to a manageable form, it
can be simplified further to the following approximation if all kt = k in Stage 1:

P0 
CF0 (1 + g 2 ) CF0 (g1 - g 2 )N
+ =
CF0
(1 + g 2 ) + N(g1 - g 2 )
(k - g 2 ) (k - g 2 ) (k - g 2 )

The intuition behind this approximation is that cash flows can be thought to grow at the eventual
constant rate of g2 from the beginning with additional “bonus” growth of (g1 – g2) for the first N
periods.

A conceptual problem with the two-stage model is that it is difficult to imagine the circumstances
under which cash flow growth would change so abruptly from g1 to g2 at a specific point in time
(i.e., Period N). A more realistic scenario is that there is a transition phase linking the end of the
first-stage cash flow growth at g1 and the final-stage cash flow growth at g2. In the three-stage
growth model, this transition period is assumed to last between Periods N1 and N2 and allows for a
linear transformation between g1 and g2. Graphically:

Cash Flow

Growth
Stage I Stage II Stage III
g1

g2

N1 N2 Time

2 - 16
The Foundations of Stock Valuation (cont.)

Formally, letting CFt represent the dollar amount of the Period t cash flow and k be the discount rate,
the value of a stock using the three-stage model can be represented as:

t - N1
 (g1 - g 2 ) 
CFN1 1 + g1 - (t - N1 )
N1
CF (1 + g1 ) t N2
 (N 2  1 - N1 )   1   CFN2 +1 
P0 =  0 t
+  t
+ N  
t =1 (1 + k) t = N1 +1 (1 + k)  (1 + k) 2   (k - g 2 ) 

= PV of “Abnormal” Growth + PV of “Transition” Growth + PV of “Constant” Growth

Notice that in this formula, the transition-period adjustment to the growth rate is designed to
calculate mid-year cash flows.

2 - 17
 Two-Stage Growth Valuation Example: Duo Growth Company

 The Duo Growth Company just paid a dividend of $1 per share. The
dividend is expected to grow at a rate of 25 percent per year for the next
three years and then to level off to 5 percent per year forever. You think
that the appropriate capitalization (i.e., discount) rate is 20 percent per
year.

 What is your estimate of the intrinsic value of a share of the stock?

 If the market price of a share is equal to this intrinsic value, what is

the expected dividend yield?

 What do you expect its price to be in one year? Is the implied capital
gain consistent with your estimate of the dividend yield and the
discount rate?

 Valuation formula for problem:

3 ($1.00)(1  0.25) t  ($1.00)(1. 25) 3 (1.05)  1 
P0     
 

t 1 (1  0.20) t  (0.20 - 0.05)  (1  0.20) 
3

2 - 18
Duo Growth Valuation Example: Solution
Projected Dividends for Duo Growth Co:

Stage 1:
Period Dividend PV(Dividend)
0 1.000 ----
1 1.250 1.042
2 1.563 1.085
3 1.953 1.130
Sum = 3.257
Stage 2:
4 2.051

Constant Growth Model (Yr 3): [2.051/(.20 - .05)] = 13.672

PV of Constant Growth: 13.672/(1.20)3 = 7.912

Current Value of Duo Growth Stock: $11.169

Expected Dividend Yield:

$1.25 / $11.17 = 11.19%

In One Year:

1.25 * (1.25) 1.25 * (1.25) 2 1.25 * (1.25) 2 1.05

P1    *  12.15
(1.2) (1.2) 2 (1.2) 2 (.2  .05)

(P1 + D1) / P0 - = (12.15 + 1.25) / 11.17 = 0.20 = k

2 - 19
 Estimating Cash Flow Growth Rates
From the preceding analysis, it should be clear that estimating cash flow growth is a major
element of the stock valuation process.

Generally speaking, there are two ways of using financial data to estimate cash flow
growth rates:
1. Historical Growth:

Let CF0 and CF-N be the per share cash flows that prevail in the current period and N periods in the past,
respectively. The compound periodic growth rate (g) that would allow CF -N to become CF0 in N periods is
defined by the formula:

CF-N x (1 + g)N = CF0

so that:

CF0
g  N -1
CF-N

That is, the historical cash flow growth rate is just the geometric average of the periodic change in the
observable cash flow levels measured at two different points in time.

2. Sustainable (i.e., Fundamental) Growth

When the cash flows involved are defined as dividend payments, a second way of estimating the long-term
growth rate is given by the following formula:

g = (Return on Equity) x (Earnings Retention Ratio) = ROE x b

where b is defined simply as [1 – Dividend Payout Ratio]. 2 - 20

 Applying the Stock Valuation Model
 The discounted cash flow approach to security valuation is the most
exhaustive method for establishing a stock’s intrinsic (i.e.,
fundamental) value. Depending on the level of confidence that the
analyst has with regard to the myriad assumptions that he or she
has made, the model implies the following trading strategy:
 If (Value) > (Market Price)  Buy Stock
 If (Value) < (Market Price)  Sell (or Short) Stock

 The valuation model also gives considerable guidance to help

analysts understand what corporate managers must do to increase
firm value:
 Increase the cash flows generated by assets in place currently
 Increase the expected growth rate of earnings
 Increase the length of the abnormal growth period
 Reduce the cost of capital that is applied to discount the cash flows

2 - 21
Applying the Stock Valuation Model: CFA Exam Question

2 - 22
Applying the Stock Valuation Model: Solution to CFA Exam Question

2 - 23
Applying the Stock Valuation Model: Solution (cont.)

2 - 24
 Applying the Stock Valuation Model:
Market-Implied Growth Rates
We have seen that with sufficient assumptions about a company’s future economic

activity—particularly with respect to the pattern of future cash flow growth—it is

possible for an analyst to estimate the firm’s intrinsic value. Of course, this quality

of this valuation process is usually quite dependent on the quality of the underlying

assumptions.

An alternative way of thinking about the valuation question is:

What growth rate of firm cash flows over the next N years would be necessary to

justify the current price of the stock?

That is, using the standard DCF model, find the value for g* assuming all other input

variables (including the current stock price) have been specified:

N

CF0 1 + g * 
t
 CFN (1 + g L ) 
  1 + k N 
-N
Current Price = + 
t =1 1 + k t 
t
 k L - g L  

One advantage of this approach is that changes the focus of the valuation exercise

from one of “guessing” about future economic conditions for the firm to one of

assessing the “reasonableness” of the growth forecast that has been priced into the

stock by the market.

2 - 25
 Implied Growth Rate Example:
Empresas COPEC - February 2005

2 - 26
COPEC Implied Growth Rate Example (cont.)

2 - 27
Model Output for COPEC Implied Growth Estimate

2 - 28
 Defining Measures of Cash Flow
It perhaps goes without saying that the discounted cash flow approach to valuing stock depends
critically on using the appropriate definition of the series of expected cash flows. There are three
definitions that are commonly used in practice:

1. Dividends

The simplest application of the discounted cash flow models involves estimating the stream of
expected dividends that the prospective shareholder will receive if he or she purchases the stock.
These dividend payments are most often stated on a per share basis:

Total Dividend Payout

Dividend Per Share  D 
Common Equity

When necessary, the common equity is usually reported on a diluted basis to account for stock
options or convertible securities that the company may have issued. The appropriate discount rate
to use with this measure of cash flow is the cost (i.e., required return) of equity.

2 - 29
 Defining Measures of Cash Flow (cont.)
2. Free Cash Flow to Equity

For many firms—particularly those with high growth opportunities—focusing on dividend

payments gives an unreliable indication of the total cash flow available to company’s shareholders
after all of the other demands on these resources (e.g., other suppliers of capital, such as
debtholders; anticipated capital expenditures) have been satisfied. A better estimate of this concept
involves the following formula:

Free Cash Flow to Equity = FCFE = (Net Income) + (Depreciation Expense)

- (Capital Expenditures) – ( Working Capital)
- (Debt Repayments) + (New Debt Issues)

Valuations based on FCFE will be identical to those using D only if the firm uses a strict residual
dividend policy (i.e., pays out whatever portion of earnings is left over after all other demands are
met). As with dividend discount models, the cost of equity is the appropriate discount rate to use
with FCFE. Finally, notice that FCFE is usually not stated on a per share basis. This means that
discounting the estimated stream of future FCFEs will yield an aggregate present equity value,
which must be divided by the current (diluted) outstanding shares to generate the per share value of
the stock.
2 - 30
 Defining Measures of Cash Flow (cont.)
3. Free Cash Flow to the Firm

Both of the preceding cash flow measures—D and FCFE—attempt to estimate the net portion of the
cash generated by the firm that is “owned” by the stockholders (after the firm’s future growth is
insured through reinvestment). With either of these measures, applying the discounted cash flow
methodology is then a straightforward matter that leads directly to the intrinsic value of the stock
share. Alternatively, the valuation mechanics can be based on the operating free cash flows
available to all of the firm’s investors (i.e., both stockholders and bondholders). The value of the
equity claim alone is established by subtracting the market value of the outstanding debt from the
intrinsic value of the entire firm. This “total” (or operating) free cash flow measure is defined as:

Free Cash Flow to Firm = FCFF = [EBIT x (1 – tax rate)] + (Depreciation Expense)
- (Capital Expenditures) – ( Working Capital)
- ( Other Assets)

where [EBIT x (1 – tax rate)] is sometimes defined as net operating profit after tax (NOPAT).
Because valuations based on FCFF involve cash flows available to all suppliers of capital, the
appropriate discount rate is the weighted average cost of capital (WACC). In practice, WACC is
frequently approximated as a weighted average of the firm’s cost of equity and it’s after-tax debt
cost, using the percentage market values of each financing sources as weights. Finally, notice that
the discounted present value of the stream of future expected FCFFs is sometimes called the
company’s enterprise value.
2 - 31
 Equity Valuation Example:
Southwest Airlines (LUV) – January 2003

 Positive analyst report in

January 2003 by S&P
Outlook
 Basis for the opinion was
the forecast of improved
growth due to cost-cutting
measures at the firm and
the company’s position in
the industry
 DCF analysis based on
both analyst forecasts and
a market-implied scenario
 Refer to Excel workbook
“LUV DCF Model” for the
details of the stock
valuation

2 - 32
LUV Stock Valuation Example (cont.)

2 - 33
LUV DCF Valuation Model Output

2 - 34
LUV DCF Valuation Model Output (cont.)

2 - 35
LUV Stock Valuation Example (cont.)

2 - 36
Valuing Special Situations: No Current Earnings
At first glance, a discounted cash flow approach to valuation would appear to be
difficult, if not impossible, to apply when the company in question may be several years
away from generating positive cash flows. Nevertheless, a modified DCF approach can
be implemented as follows:

Step 1: Choose a set of assumptions about how the firm will look at some point in the
future and use this projection to value the firm at that time using a variation of the
standard DCF model:

N CF0 [  1 + g t  ]  CFN (1 + g 2 ) 
  1 + k N 
-N
Value0 = + 
t =1 1 + k t 
t
 k N - g 2  

where CFt is the projected cash flow for period t, gt is the period t cash flow growth rate,
and kN is the relevant cost of capital.

- For example, it may be five years before you think the company will generate revenues
sufficient to cover its operating expenses and debt service, as well as generate
predictable future growth.

Step 2: Repeat Step 1 with several alternative sets of assumptions about what the firm
may look like in the future, thereby creating a “distribution” of potential future values.
2 - 37
 Valuing Special Situations (cont.)
- Alternative assumptions could affect company fundamentals such as revenue
growth, operating margin, or debt ratios. Once generated, these alternative forecasts
might be labeled as “optimistic”, “neutral”, and “pessimistic”.

Step 3: All of the projected future values should be discounted back to the present,
using the relevant cost of capital statistic.

Step 4: Probabilities can be assigned to the likelihood of each potential future

scenario. These probabilities can then be used in establishing an expected present
value for the company by calculated a weighted average of the discounted future
values.

- That is, if pi represents the estimated probability of the i-th scenario, the expected
value of the firm can be estimated:
Expected Present Value = p1 x PV(Value)1 + ….. + pm x PV(Value)m

2 - 38
Valuing a Negative EPS Company: AMZN in January 2001

2 - 39
Valuing a Negative EPS Company: AMZN in January 2001 (cont.)

2 - 40
Overview of Comparable Multiples Approach to Valuation
The underlying assumption of a “comparables” approach to equity valuation is that
you need to be fully invested in the market at the current time. That is, the question
facing the investor is which stock should be held, rather than whether any stock
should be held.

Some of the more popular relative valuation metrics used in practice include:

Measure Comment

Price/Earnings The most widely used comparable; easy to

compute based on reported or forecast EPS

Price/Cash Flow Less subject to accounting manipulation;

particularly useful for companies without
earnings (e.g., oilfield service, E&P)

Price/Sales Based on “top line” revenue; used in turn-

around situations and start-up companies
with EPS or FCF farther out in the
future (e.g., dot.coms)

Price/Book Value Used most frequently to define the “value”

vs. “growth” style of investing

Enterprise Value/ Consistent with the FCFF” DCF valuation

EBITDA approach; used with highly levered firms in
capital intensive industries (e.g.,
entertainment, LBO/takeover)

PEG Ratio The Price/Earnings ratio divided by annual

earnings growth; a cost-benefit measure
calculated on historical or forecast basis 2 - 41
Relationship Between DCF and Comparable Multiple Valuation Approaches
Analysts attempting to establish the fundamental value of a particular stock often find themselves
choosing between one of two different approaches. First, they might consider using the discounted
cash flow (DCF) technique, which seeks to link the value of a company directly to its expected
future cash flows and discount rates. Conversely, analysts can try to measure the firm’s relative
value by focusing on a series of valuation multiples that tie the current stock price to any of several
accounting measures (e.g., Price/Earnings, Price/Book, Price/Sales).

Although DCF and relative valuation methods are often perceived as “competitors”—sometimes to
the point where analysts identify exclusively with one technique or the other—there is a conceptual
connection between them. In fact, as the following discussion reveals, they can really be thought of
as deriving from the same common foundation.

To see this connection, we start with the constant-growth dividend discount model, which is the
simplest form of the DCF approach. Letting Dt be the period t dividend per share, k be the cost of
equity, and g be the constant annual growth rate of dividends (and further assuming that k > g), we
have:

 
E(D t ) D 0 (1  g) t
P0   (1  k) t
  (1  k) t
t 1 t 1

D1

(k - g)

Notice that this version of the DCF model reduces a stock’s predicted value to three variables: next
period’s expected dividend, the constant discount rate, and the constant dividend growth rate.

Consider now how this basic form of the DCF model relates to several metrics used in assessing a
stock’s relative value:

1. Dividend Yield

The dividend yield is the ratio of next period’s expected dividend to the current stock price (i.e.,
D1/P0). Using the basic DCF equation of value as a surrogate for the current price, we have:

(D1/P0) = (k – g)

Notice one consequence of this relationship is that if (k – g) represents the stock’s dividend yield,
then g must represent the rate at which the stock’s price is expected to appreciate. This is because
the stock’s expected return (i.e., k) can be viewed as the sum of (i) the expected capital gain and (ii)
the expected cash payout to the investor.
2 - 42
DCF and Comparable Multiple Valuation Approaches (cont.)
2. Price/Earnings Ratio

To see the connection between the DCF approach and the Price/Earnings multiple—which is
arguably the most commonly used relative valuation metric in practice—first define the company’s
expected dividend payout ratio as:

d = (D1/E1)

where E1 is the expected level of next period’s earnings per share. The forward Price/Earnings ratio
(i.e., based on forecasted earnings) can then be written as:

P0 (D1 /E 1 ) d
 
E1 (k - g) (k - g)

Notice that this formula suggests that the level of the valuation multiple is directly related to the
company’s sustainable dividend payout policy, which itself is a function of the firm’s long-term
growth potential. Also, the spread between k (which is driven by the company’s systematic level of
risk) and g is inversely related to the size of the P/E ratio.

Although the level of d clearly impacts the P/E ratio, it is best to consider this variable as the firm
management’s long-run target payout policy, which should be relatively stable over time. Thus, the
main determinant of the earnings multiple in this framework remains the long-term growth potential
of the company.

3. Price/Book Ratio

In a similar manner to that just shown, the DCF model estimate of the company’s present value can
be linked to its book value (BV) as follows:

P0 (D1 / BV0 ) (d x [E1 / BV0 ])

 
BV0 (k - g) (k - g)

(d x ROE 1 )

(k - g)

where ROE1 is the firm’s expected return on equity and d continues to be the expected dividend
payout.

The most important thing to notice from this formulation is the Price/Book multiple is positively
related to future company performance. Specifically, notice that P/BV is directly related to next
period’s ROE. This suggests once again that to be useful, a valuation multiple must be forward
looking. Also, because ROE can itself be decomposed further (e.g., by the DuPont method, ROE =
(E/S) x (S/A) x (A/BV) = [Profit Margin] x [Asset Turnover] x [Financial Leverage]), analysts can 2 - 43
use the Price/Book multiple to further refine their understanding of how value is created in the firm.
DCF and Comparable Multiple Valuation Approaches (cont.)
4. Price/Sales Ratio

Rather than focus exclusively on the “bottom line,” analysts often also attempt to tie a company’s
valuation to its top-line sales revenue. Letting next period’s sales per share be expressed as S 1, the
Price/Sales multiple can be tied to the DCF framework as follows:

P0 (D1 /S1 ) (d x [E1/S1 ])

 
S1 (k - g) (k - g)

(d x [Net Profit Margin] 1 )


(k - g)

where the net profit margin is again a forward-looking measure. Although such forecasts are not
always widely available, the company’s Price/Sales ratio is seen to be directly related to the
company’s ability to sustain increases in its net margin.

In summary, the preceding discussion underscores the point that there is a tractable foundation for
the relationship between the DCF and relative approaches to security valuation. In particular, the
various relative valuation metrics each have a key performance variable that serves as the main
driver for value creation. These connections can be summarized as follows:

Relative Metric Key Determinant Other Determinants

Dividend Yield (D/P) Long-Term Growth (g) k

Price/Earnings (P/E) Long-Term Growth (g) k, d
Price/Book (P/BV) Return on Equity (E/BV) k, d, g
Price/Sales (P/S) Net Profit Margin (E/S) k, d, g
2 - 44
Using Comparable Multiples in Practice: CFA Exam Question

2 - 45
Solution to CFA Exam Question

2 - 46
Solution to CFA Exam Question (cont.)

2 - 47
 Using Comparable Multiples in Security Valuation
 Asset-based valuation multiples (i.e., those using book value) generally
produce smaller valuation errors than those using sales (from Lie and Lie,
Financial Analysts Journal, 2002)

2 - 48
Comparable Multiple Valuation Example: LUV

2 - 49
Comparable Multiple Valuation Example: LUV (cont.)

2 - 50
Comparable Multiple Valuation Example: LUV (cont.)

2 - 51
LUV Earnings Forecast Model: Bear Stearns – January 2005

2 - 52
LUV Stock Recommendation: Bear Stearns – February 2005

2 - 53
Comparable Multiple Valuation Example: COPEC – February 2005

2 - 54
Comparable Multiple Valuation Example: COPEC (cont.)

2 - 55
Comparable Multiple Valuation Example: COPEC (cont.)

2 - 56
Overview of Equity Portfolio Management Strategies

• Passive Management Strategies

1. Efficient Markets Hypothesis

- Buy-and-Hold
- Indexing

• Active Mangement Strategies

2. Fundamental Analysis
- “Top Down” (e.g., asset class rotation, sector rotation)
- “Bottom Up” (e.g., stock undervaluation/overvaluation)

3. Technical Analysis
- Contrarian (e.g., overreaction)
- Continuation (e.g., price momentum)

4. Anomalies and Attributes

- calendar effects (e.g., Weekend, January)
- security characteristics (e.g., P/E, P/B, earnings momentum, firm size)
- investment style (e.g., value, growth)
2 - 57
 An Efficient Capital Market
 A capital market is considered to be efficient if, through their trading
activities, investors set the price of any particular security in a manner that
impounds new information about that security in an instantaneous manner.

 Said differently, an efficient market is one in which all security prices are set
as if all available information has already been assimilated by investors and
traders and that information has been acted upon in the proper way. Thus,
the only thing that will change the security’s market price is the arrival of
new information which, by definition, is not fully predictable.

 Notice from the preceding discussion that the critical concept defining an
efficient market is not if new information about a particular security is
reflected in the security’s market price, but how rapidly the price adjusts to
this new information.

 In establishing whether capital markets are efficient, it is often useful to

consider the nature of the information that the market is expected to react
to:
 Weak Form Efficiency: Information contained in past price movements only.
 Semi-Strong Form Efficiency: Public information announcements (e.g., earnings
announcements, corporate restructurings)
 Strong Form Efficiency: Non-public information (e.g., insider trading)

2 - 58
Efficient vs. Inefficient Information Processing

2 - 59
 Market Efficiency: Implications and Evidence
 One direct implication of capital markets that are economically (if not
perfectly) efficient is that it will be impossible over time for a money
manager to consistently add “alpha” to a client’s portfolio through such
activities as market timing or superior stock selection.

 This in turn suggests that a passive indexing of asset class investments with
the appropriate risk level is the appropriate strategy to follow.

 Empirical research on capital market efficiency has established the following

stylized “facts”:
 Markets are generally efficient in both the weak and semi-strong forms over time,
but there are some important and consistent deviations from this rule.
 Markets are generally not strong form efficient, but the number of people who
genuinely possess inside information is smaller than those who think they do.
 It is very difficult to establish market efficiency without specifying a model for
expected returns (e.g., CAPM, Fama-French three-factor model). This means
that any conclusions about market efficiency are subject to the possibility that the
expected return model was mis-specified. (This is sometimes referred to as the
joint hypothesis problem.)

2 - 60
 Two Important Market Efficiency “Anomalies”

 Market Overreaction

2 - 61
 Two Important Market Efficiency “Anomalies” (cont.)

 Market Underreaction (i.e., Momentum)

2 - 62
 Active Equity Management: Technical vs. Fundamental Approaches
Technical Approaches:

 A contrarian investment strategy is based on the belief that the best time to
buy (sell) a stock is when the majority of other investors are the most
bearish (bullish) about it. In this way, the contrarian investor will attempt to
always purchase the stock when it is near its lowest price and sell it (or
even short sell it) when it nears its peak. Implicit in this approach is the
belief that stock returns are mean-reverting, indicating that over time stocks
will be priced so as to produce returns consistent with their risk-adjusted
expected (i.e., mean) returns. The overreaction hypothesis shows that
investing on this basis can provide consistently superior returns.

 At the other extreme, active portfolios can also be formed on the

assumptions that recent trends in past prices will continue. A price
momentum strategy, as it is more commonly called, assumes that stocks
that have been hot will stay hot, while cold stocks will also remain so.
Although there may well be sound economic reasons for these trends to
continue (e.g., company revenues and earnings that continue to grow faster
than expected), it may also simply be the case that investors periodically
underreact to the arrival of new information. Thus, a pure price momentum
strategy focuses just on the trend of past prices alone and makes purchase
and sale decisions accordingly.
2 - 63
 Active Equity Management: Technical vs. Fundamental Approaches (cont.)

Fundamental Approaches:

 An earnings momentum strategy is a somewhat more formal active portfolio

approach that purchases and holds stocks that have “accelerating” earnings and sells
(or short sells) stocks with disappointing earnings. The notion behind this strategy is
that ultimately a company’s share price will follow the direction of its earnings, which
is one “bottom line” measure of the firm’s economic success. In judging the degree
of momentum in a firm’s earnings, it is often the case in practice that investors will
compare the company’s actual EPS to some level of what was expected. Two types
of expected earnings are used most frequently: (i) those generated by a statistical
model and (ii) the consensus forecast of professional stock analysts. The previous
chart shows that over the 1994-1998 period earnings momentum strategies were
generally successful as well, although surprisingly not to the same degree as price
momentum strategies.

 A more promising approach to active anomaly investing involves forming portfolios

based on various characteristics of the companies themselves. Two characteristics
that consistently matter in the stock market are the total capitalization of the firm’s
outstanding equity (i.e., firm size) and the financial position of the firm, as indicated
by its various financial ratios (e.g., P/E, P/BV). Both attributes are commonly used to
define the nature of style investing. There are two general conclusions we can
make about these firm characteristics. First, over time, firms with smaller market
capitalizations produce different risk-adjusted returns than those with large market
capitalizations. Second, over time, firms with lower P/E and P/BV ratios (i.e., value
stocks) produce bigger risk-adjusted returns than those with higher levels of those
ratios (i.e., growth stocks).
2 - 64
Equity Portfolio Strategy Example: HACAX

2 - 65
 Equity Portfolio Strategy Example: HACAX (cont.)
Investment Philosophy
 Stocks must demonstrate superior absolute and relative earnings growth and be attractively valued relative
to expectations
 Larger capitalization growth stocks
 Characteristics generally demonstrated by portfolio companies include
 superior sales growth - improving sales momentum
 high level of unit growth - the true measure of a growth company
 high or improving ROE and ROA
 strong market position with a defensible franchise
 strong balance sheet
 some distinctive attributes such as
 unique marketing competence
 strong R&D, resulting in a superior new product flow
 excellent management capability including financial discipline
 earnings progression is of uppermost importance
 prefer companies in the early stages of exhibiting these characteristics
 perfect stock is one that Wall Street thinks will grow at 14%, Jennison thinks will grow at 18% and
it actually grows at 25%
 Investment philosophy is clearly focused and closely adhered to; there is little deviation
 All professionals are paid on the basis of their effect on the firm
 Sell Criteria
 sell if growth expectations are not achieved or exceeded
 reduce holding if a stock gets ahead of itself or reaches its price target
 lighten the position if something is not going right - if things continue to go wrong sell some more;
buy back if things straighten out
 buy a new holding if better than weakest holding; sell a position before buying a new one
 fundamentals deteriorate; eliminate if original premise is no longer present
 Frequently add to a position on price weakness attributable to a non-operating problem
 1% minimum and 5% maximum individual stock position - most positions are 1% to 3% (those less than
1% are either in the early stages of being assembled, or are in the final stages of being eliminated)
 Size of individual stock positions is based on confidence in the company and its management
 Bottom-up stock selection
 Almost all research is done internally; this is the most important component of Jennison's active
management process
 Do not buy a stock without visiting the company 2 - 66
 Equity Portfolio Strategy Example: HACAX (cont.)

Investment Risks
Stocks do fluctuate in price and the value of your investment in the fund may go down. This means that you could lose money on your
investment in the fund or the fund may not perform as well as other possible investments if any of the following occurs:

 A drop in U.S. or foreign stock markets.

 The market favors small cap stocks over medium and large cap stocks, or value over growth stocks.
 An adverse event, such as unfavorable earnings report, depresses the value of a particular company’s stock.
 The subadviser’s judgment about the attractiveness, value and potential appreciation of particular companies’ stocks prove to be incorrect.
 Prices of the fund’s foreign securities go down because of unfavorable changes in foreign currency exchange rates, foreign government actions,
political instability or the more limited availability of accurate information about foreign issuers. These risks are more severe for issuers in emerging
market countries.

The fund's performance may be more volatile because it invests in mid cap stocks. Mid cap companies may have more limited product lines,
markets and financial resources than large cap companies. They may also have shorter operating histories and more volatile businesses. Mid
cap stocks tend to trade in a wider price range than large cap stocks. In addition, it may be harder to sell these stocks, particularly in large
blocks, which can reduce their selling price.

2 - 67
Equity Portfolio Strategy Example: HACAX (cont.)

2 - 68
Equity Portfolio Strategy Example: HACAX (cont.)

2 - 69
 Active vs. Passive Equity Portfolio Management
 The “conventional wisdom” held by many investment analysts is that
there is no benefit to active portfolio management because:
 The average active manager does not produce returns that exceed
those of the benchmark
 Active managers have trouble outperforming their peers on a consistent
basis

 However, others feel that this is the wrong way to look at the Active
vs. Passive management debate. Instead, investors should focus
on ways to:
 Identifying those active managers who are most likely to produce
superior risk-adjusted return performance over time

 This discussion is based on research authored jointly with Van

Harlow of Fidelity Investments titled:
“The Right Answer to the Wrong Question:
Identifying Superior Active Portfolio Management”

2 - 70
 The Wrong Question

Stylized Fact:

Most active mutual fund managers cannot outperform the S&P 500
index on a consistent basis
Beat %
90%

70%

50%

30%

10%
JAN80 JAN82 JAN84 JAN86 JAN88 JAN90 JAN92 JAN94 JAN96 JAN98 JAN00 JAN02 JAN04
DATE

2 - 71
 Defining Superior Investment Performance

 Over time, the “value added” by a portfolio manager

can be measured by the difference between the
portfolio’s actual return and the return that the portfolio
was expected to produce.

 This difference is usually referred to as the portfolio’s

alpha.

Alpha = (Actual Return) – (Expected Return)

2 - 72
 Measuring Expected Portfolio Performance
 In practice, there are three ways commonly used to measure the return
that was expected from a portfolio investment:

 Benchmark Portfolio Return

 Example: S&P 500 or Russell 1000 indexes for a U.S. Large-Cap Blend fund
manager
 Pros: Easy to identify; Easy to observe
 Cons: Hypothetical return ignoring taxes, transaction costs, etc.; May not be
representative of actual investment universe; No explicit risk adjustment

 Peer Group Comparison Return

 Example: Median Return to all U.S. Small-Cap Growth funds for a U.S. Small-Cap
Growth fund manager
 Pros: Measures performance relative to manager’s actual competition
 Cons: Difficult to identify precise peer group; “Median manager” may ignore large
dispersion in peer group universe; Universe size disparities across time and fund
categories

 Return-Generating Model
 Example: Single Risk-Factor Model (CAPM); Multiple Risk-Factor Model (Fama-
French Three-Factor, Carhart Four-Factor)
 Pros: Calculates expected fund returns based on an explicit estimate of fund risk;
Avoids arbitrary investment style classifications
 Cons: No direct investment typically; Subject to model misspecification and factor
measurement problems; Model estimation error
2 - 73
 The Wrong Question (Revisited)
Stylized Fact:

Across all investment styles, the “median manager” cannot produce
positive risk-adjusted returns (i.e., PALPHA using return model)

Monthly Mean PALPHA Value at Percentile (%):

% Pos.
Fund Style # of Obs. 5th 25th Median 75th 95th Alphas
Overall 19551 -1.56 -0.55 -0.18 0.12 0.79 33.77
LV 2,387 -2.11 -0.57 -0.21 0.07 0.66 23.51
LB 3,377 -1.44 -0.55 -0.22 -0.01 0.38 42.02
LG 3,351 -1.08 -0.38 -0.07 0.17 0.80 30.21
MV 1,413 -2.61 -0.67 -0.23 0.11 0.69 29.10
MB 1,691 -1.86 -0.79 -0.32 0.07 0.64 35.31
MG 3,169 -1.48 -0.63 -0.21 0.19 1.04 32.77
SV 929 -2.02 -0.65 -0.25 0.01 0.57 32.16
SB 1,222 -1.42 -0.59 -0.19 0.12 0.77 48.46
SG 2,012 -1.37 -0.45 -0.02 0.39 1.24 25.62
S&P 500 0.04
Index Fund

2 - 74
 The Right Answer

 When judging the quality of active fund managers, the important

question is not whether:
 The average fund manager beats the benchmark
 The median manager in a given peer group produces a positive
alpha

 The proper question to ask is whether you can select in advance

those managers who can consistently add value on a risk-adjusted
basis
 Does superior investment performance persist from one period to the
next and, if so, how can we identify superior managers?

2 - 75
 Lessons from Prior Research
 Fund performance appears to persist over time
Original View:

Managers with superior performance in one period are equally likely to produce superior or inferior performance in
the next period


Current View:
Some evidence does support the notion that investment performance persists from one period to the next

The evidence is particularly strong that it is poor performance that tends to persist (i.e., “icy” hands vs. “hot” hands)

Securitycharacteristics, return momentum, and fund style appear to influence fund

performance

 Security Characteristics:
After controlling for risk, portfolios containing stocks with different market capitalizations, price-earnings ratios, and
price-book ratios produce different returns

Funds with lower portfolio turnover and expense ratios produce superior returns


Return Momentum:
Funds following return momentum strategies generate short-term performance persistence

When used as a separate risk factor, return momentum “explains” fund performance persistence

2 - 76
 Lessons from Prior Research (cont.)

Security characteristics, return momentum, and fund style appear to influence fund
performance (cont.)
 Fund Style Definitions:
After controlling for risk, funds with different objectives and style mandates produce different returns

Value funds generally outperform growth funds on a risk-adjusted basis


Style Investing:
Fund managers make decisions as if they participate in style-oriented return performance “tournaments”

The consistency with which a fund manager executes the portfolio’s investment style mandate affects fund
performance, in both up and down markets

Active fund managers appear to possess genuine investment skills


Stock-Picking Skills:
Some fund managers have security selection abilities that add value to investors, even after accounting for fund
expenses

A sizeable minority of managers pick stocks well enough to generate superior alphas that persist over time


Investment Discipline:
Fund managers who control tracking error generate superior performance relative to traditional active managers
and passive portfolios

 Manager Characteristics:
The educational backgrounds of managers systematically influence the risk-adjusted returns of the funds they
manage
2 - 77
 Data and Methodology for Performance Analysis
 CRSP (Center for Research in Security Prices) US Mutual Fund
Database
 Survivor-Bias Free database of monthly returns for mutual funds for the period 1962-2003

 Screens
Diversified domestic equity funds only
Eliminate index funds
Require 30 prior months of returns to be included in the analysis on any given date
Assets greater than $1 million
Period 1979 – 2003 in order to analyze performance versus an index fund and have sufficient
number of mutual funds

 Return-generating model:
Fama-French
E(Rp) = RF + {bm[E(Rm) – RF] + bsml[SML] + bhml[HML]}

 Style classification
 Map funds to Morningstar-type style categories based on Fama-French SML and HML
factor exposures (LV, LB, LG, MV, MB, MG, SV, SB, SG) 2 - 78
 Methodology: Fund Mapped by Style Group

Mutual Fund Style Category:

Year LV LB LG MV MB MG SV SB SG Total

1979 9 23 70 0 3 21 0 3 27 156
1980 7 26 59 2 7 36 2 3 30 172
1981 5 20 32 1 6 39 0 3 25 131
1982 13 23 38 1 5 43 0 1 34 158
1983 14 27 60 1 7 31 0 2 42 184
1984 8 26 55 1 1 37 0 4 35 167
1985 7 23 74 3 1 37 7 1 30 183
1986 5 18 95 3 5 41 12 0 18 197
1987 6 22 80 3 2 51 14 5 16 199
1988 9 29 89 3 8 50 10 7 31 236
1989 12 30 92 2 3 54 0 11 43 247
1990 19 42 92 1 3 46 1 3 53 260
1991 25 63 97 3 3 44 0 2 48 285
1992 32 163 176 7 10 77 3 11 90 569
1993 38 202 166 8 5 92 2 19 103 635
1994 49 269 198 4 16 148 3 24 162 873
1995 57 210 224 20 67 234 24 97 264 1197
1996 86 405 421 20 45 279 47 83 262 1648
1997 160 535 478 52 111 357 83 106 324 2206
1998 355 636 601 160 130 256 133 172 356 2799
1999 469 456 827 157 107 641 261 119 412 3449
2000 771 604 992 316 82 587 215 142 459 4168
2001 812 680 1181 302 129 699 155 110 457 4525
2002 907 962 840 345 193 835 99 194 647 5022
2003 836 1250 1078 226 375 764 242 263 580 5614
2 - 79
 Methodology (cont.)

Evaluate
Estimate Model Performance

Time
36 Months 3 Months (1 Month)

 Use past 36 months of data to estimate model parameters

 Standardized data within each peer group on a given date to allow for time-
series and cross-sectional pooling [Brown, Harlow, and Starks (JF, 1996)]

 Evaluate performance
 Use estimated model parameters to calculate out-of-sample alphas based on
factor returns from the evaluation period

 Roll the process forward one quarter (one month) and

estimate all parameters again, etc.
2 - 80
 Performance Analysis
Distributions of Out-of-Sample Future Alphas (FALPHA)
Quarterly – Equally Weighted 1979-2003

Quarterly FALPHA Value at Percentile (%):

% Pos.
Fund Style # of Obs. 5th 25th Median 75th 95th Alphas
Overall 126,613 -8.85 -3.12 -0.49 2.06 8.55 44.50
LV 17,195 -7.53 -2.98 -0.66 1.82 6.80 42.28
LB 23,566 -7.07 -2.43 -0.48 1.28 6.10 42.37
LG 30,642 -7.95 -2.66 -0.25 1.89 7.99 46.59
MV 6,214 -10.82 -3.13 -0.09 2.93 9.41 49.10
MB 4,251 -8.21 -3.23 -0.24 2.88 9.06 47.49
MG 19,172 -9.71 -3.79 -0.56 2.67 10.32 45.34
SV 4,963 -12.37 -4.39 -1.30 1.99 10.81 38.32
SB 4,475 -9.95 -3.96 -1.12 1.89 8.47 40.20
SG 16,135 -11.07 -4.03 -0.59 3.10 10.89 45.53
S&P 500 295 -1.41 -0.37 0.08 0.51 1.22 54.58
Index Fund

2 - 81
 Time Series Analysis

Pooled Regressions – Fund Characteristics versus Future Alpha

1979-2003

1 Month Alpha 3 Month Alpha

Parameter Parameter
Variable Estimate Prob Estimate Prob
Intercept 0.000 1.000 0.000 1.000
Past Alpha 0.071 0.000 0.072 0.000
Expense Ratio (0.012) 0.000 (0.023) 0.000
Diversify (R-Sq) (0.036) 0.000 (0.055) 0.000
Volatility (0.012) 0.000 (0.006) 0.043
Turnover 0.016 0.000 0.019 0.000
Assets 0.007 0.000 0.008 0.009

2 - 82
 Cross-Sectional Analysis

 Use past 36 months of data to estimate model parameters

 Run a sequence of Fama-MacBeth cross-sectional regressions of
future performance against fund characteristics and model
parameters (alpha and R2 )
 Average the coefficient estimates from regressions across the
entire sample period
 T-statistics based on the time-series means of the coefficients

2 - 83
 Cross-Sectional Performance Results

Fama-MacBeth Regressions – Fund Characteristics versus Future Alpha

1979-2003

1 Month Alpha 3 Month Alpha

Parameter Parameter
Variable Estimate Prob Estimate Prob

Past Alpha 0.047 0.000 0.061 0.000

Expense Ratio (0.012) 0.033 (0.019) 0.063
Diversify (R-Sq) (0.021) 0.091 (0.023) 0.333
Volatility (0.011) 0.377 (0.022) 0.306
Turnover 0.015 0.034 0.022 0.072
Assets 0.008 0.034 0.009 0.190

2 - 84
 Logit Performance Analysis

Fund Characteristics versus a Positive Future Alpha

1979-2003

1 Month Alpha 3 Month Alpha

Parameter Parameter
Variable Estimate Prob Estimate Prob
Intercept (0.159) 0.000 (0.228) 0.000
Past Alpha 0.082 0.000 0.093 0.000
Expense Ratio (0.021) 0.000 (0.033) 0.000
Diversify (R-Sq) (0.085) 0.000 (0.117) 0.000
Volatility (0.003) 0.419 (0.022) 0.000
Turnover 0.028 0.000 0.022 0.000
Assets 0.015 0.000 0.023 0.000

2 - 85
 Probability of Finding a Superior Active Manager

Probability of Future Positive 3-month Alpha

Median Manager Controls for Turnover, Assets, Diversify, and Volatility

EXPR:

Std. Dev. -2 (Low) -1 0 +1 +2 (High) (High –

Group Low)

-2 (Low) 0.4143 0.4062 0.3982 0.3903 0.3824 (0.0319)

-1 0.4369 0.4288 0.4206 0.4125 0.4045 (0.0324)

PALPHA: 0 0.4599 0.4516 0.4434 0.4352 0.4270 (0.0329)

+1 0.4830 0.4746 0.4664 0.4581 0.4498 (0.0331)

+2 (High) 0.5061 0.4978 0.4895 0.4812 0.4729 (0.0333)

(High – 0.0918 0.0916 0.0913 0.0909 0.0905

Low)

2 - 86
Probability of Finding a Superior Active Manager (cont.)

Probability of Future Positive 3-month Alpha

“Best” Manager Controls for Turnover, Assets, Diversify, and Volatility

EXPR:

Std. Dev. -2 (Low) -1 0 +1 +2 (High) (High – Low)

Group

PALPHA: -2 (Low) 0.5051 0.4968 0.4884 0.4801 0.4718 (0.0333)

-1 0.5282 0.5199 0.5116 0.5033 0.4950 (0.0333)

0 0.5512 0.5430 0.5347 0.5264 0.5181 (0.0331)

+1 0.5741 0.5659 0.5577 0.5495 0.5412 (0.0328)

+2 (High) 0.5965 0.5885 0.5804 0.5723 0.5641 (0.0324)

(High – Low) 0.0915 0.0918 0.0920 0.0922 0.0923

2 - 87
Portfolio Strategies Based on Active Manager Search

Asset Weighted Alpha Deciles - Quarterly Rebalance

1979-2003
2.00%

1.50%

1.00%
Average Annualized Alpha

0.50%

0.00%
1 2 3 4 5 6 7 8 9 10
-0.50%

-1.00%

-1.50%

-2.00%

-2.50%
2 - 88
 Portfolio Strategies (cont.)

Asset Weighted - Quarterly Rebalance

Formation Variables Separated by Upper and Lower Quartile Values
1979-2003

Portfolio Formation Variables

Return
Cumulative Value Average Alpha Alpha Volatility Differential
Expense Alpha of $1 Invested (%) (%) (bp)

Overall Sample 1.046 0.181 2.153

Lo 1.009 0.037 2.142 2

Hi 1.005 0.018 4.025
Hi 1.515 1.691 3.371 291
Lo 0.738 (1.221) 3.469
Lo Hi 1.446 1.502 3.596 309
Hi Lo 0.673 (1.585) 4.712

S&P 500 Index Fund 1.022 0.088 1.700

2 - 89
Implementing a “Fund of Funds” Strategy: An Example
Methodology
Evaluate
Estimate Model Performance

Time
9 Months 3 Months (1 Month)

 Use past 9 months of daily data to estimate model and in-

sample alpha

 Optimize portfolio based on an assumption of risk aversion,

i.e., risk-return tradeoff preference

 Compute the performance of the portfolio over the next three

(one) months

 Roll the process forward each quarter and estimate all

parameters again, etc. 2 - 90
 “Fund of Funds” Strategy
Fidelity Advisor Diversified Equity Fund Styles (6/04)

100 FAIVX IVV FGIOXFDGIX FHCIX

EQPIX
FFSIX FAGCX
90 FUGIX
FALIX
Cap: Small to Large

80 FCNIX FTQIX

FCLIX
70 FTIMX FATIX
FHEIX EQPGX

60 FFYIX

50
FMCCX
FRVIX
40

FSCIX FASOX
30 FBTIX
FDCIX

20

10

0 FVLIX
FVIFX

0 10 20 30 40 50 60 70 80 90 100

Value to Grow th 2 - 91
 “Fund of Funds” Portfolio Strategy
Portfolio Weights Over Time
Name 200103 200106 200109 200112 200203 200206 200209 200212 200303 200306 200309 200312 200403
Fidelity Advisor Equity Growth Instl 7.1% 5.3% 9.9% 9.6%
Fidelity Advisor Equity Income Instl 20.0% 20.0% 20.0% 19.5% 20.0% 20.0% 12.0% 9.0% 5.3% 20.0%
Fidelity Advisor Growth Opport Instl 13.4% 10.0% 2.6% 2.6% 10.8% 10.6% 14.5% 14.6% 6.1% 2.9%
Fidelity Advisor Equity Value I 18.2% 18.6% 18.5% 18.8% 18.5% 10.4% 10.6% 10.6%
Fidelity Advisor Large Cap Instl 7.0% 7.0%
Fidelity Advisor Value Strat Instl 0.3% 4.5% 4.6% 4.9%
Fidelity Advisor Technology Instl 8.0% 8.1% 7.5% 6.1% 5.6% 4.8% 4.2% 6.2% 5.0% 6.0% 7.3% 7.3% 5.6%
Fidelity Advisor Cyclical Indst Instl 6.0% 7.4% 0.7% 4.9% 3.9% 4.1% 0.7% 1.2% 1.2% 10.9% 11.1%
Fidelity Advisor Consumer Indst Instl 2.5% 1.4% 1.5% 2.5% 1.7%
Fidelity Advisor Dynamic Cap App Inst 6.1% 9.3% 0.5%
Fidelity Advisor Dividend Growth Inst 20.0% 20.0% 20.0% 20.0% 20.0% 2.1% 2.1% 7.6% 15.8% 15.9% 4.3% 4.2% 4.2%
Fidelity Advisor Financial Svc Instl 9.3% 8.6% 7.8% 7.1% 1.2% 1.8% 11.8% 11.7% 11.5% 4.7% 12.3% 12.4% 15.1%
Fidelity Advisor Growth & Income Inst 12.1% 16.7% 20.0% 17.1% 15.6% 7.4% 7.3% 11.1% 17.5% 17.4% 5.8% 5.7%
Fidelity Advisor Health Care Instl 3.2% 2.2% 4.7% 4.5% 4.0% 8.8% 10.8% 9.7% 6.0% 5.7% 5.5% 5.5%
Fidelity Advisor Mid Cap Instl 2.3% 9.6% 10.1% 8.6%
Fidelity Advisor Telecomm&Util Gr Ins 0.9% 0.9% 5.0% 1.9% 5.0% 4.7% 4.7% 1.4%
iShares S&P 500 Index 11.7% 15.2% 20.0% 7.8% 8.2% 8.1% 12.8% 14.9% 8.6% 8.5% 8.5% 12.3% 15.0%

Portfolio Characteristics
Avg Annual % Beat % Beat Worst Longest Longest Annual
Active Bench in Bench in Best Active Active Winning Losing Tracking
Portfolio Return Periods Up Market Down Market Return Return Streak Streak Error

S&P 500 0.68% 84 60% 67% 5.20% (3.7%) 7 4 3.3%

2 - 92
 Cumulative Returns versus S&P 500
0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

-0.1
JAN97 JAN98 JAN99 JAN00 JAN01 JAN02 JAN03 JAN04 JAN05

2 - 93
 Active vs. Passive Management: Conclusions

 Both passive and active management can play a

role in an investor’s portfolio

 Strong evidence for both positive and negative

performance persistence (i.e., alpha persistence)
 Prior alpha is the most significant variable for forecasting future alpha
 Expense ratio, risk measures, turnover and assets are also useful in
forecasting future alpha

 The existence of performance persistence provides

a reasonable opportunity to construct portfolios
that add value on a risk-adjusted basis

2 - 94
 Overview of the Hedge Fund Industry
 Generally speaking, a hedge fund is an organizational structure for
managing private investment capital in a relatively unrestricted manner.
Unlike the mutual fund industry we have just studied—which typically
imposes severe restrictions on investment activities (e.g., short sale
constraints, leverage restrictions)—hedge funds generally face fewer or no
such restrictions.

 Hedge funds are usually classified in the alternative asset category in a

portfolio’s strategic asset allocation.

 Formally, a hedge fund is a managed portfolio that attempts to preserve

invested capital, reduce volatility, and provide positive returns under all
market conditions. Hedge fund managers attempt to accomplish these
goals by taking long and short positions in various securities, using leverage
and derivatives, employing arbitrage strategies, and taking positions in
virtually any security in which a superior return opportunity is attainable.

 Sometimes hedge funds are categorized under the more generally heading
of absolute return investment strategies because they seek to provide
investors with positive returns regardless of the direction of general market
movements. However, it is important to recognize that there are a wide
variety of investment strategies under the hedge fund umbrella,
representing a significant range of the investment risk spectrum. 2 - 95
 Overview of the Hedge Fund Industry (cont.)

 The global hedge fund

industry has grown quite
rapidly over the past decade.
From about 1,000 funds at the
start of the 1990s controlling
less than $50 billion in assets,
by 2004 there were almost
9,000 active funds controlling
an estimates $975 billion in
assets.

 The adjacent charts

summarize the rapid growth in
the assets of the hedge fund
industry, which have been
growing at about 20% per
annum in recent years.
2 - 96
Overview of the Hedge Fund Industry (cont.)

2 - 97
 Overview of Hedge Fund Strategies
 Saying you manage a hedge fund is like saying you play a sport;
that comment offers some information, but it is not very specific. In
practice, there are several different broad categories of hedge fund
investment strategies:

 Equity-Based Strategies
 Long-Short Equity: Perhaps the most “basic” form of hedge fund
investing, managers attempt to identify misvalued stocks and take long
positions in the undervalued ones and short positions in the overvalued
ones. Given that managers may participate in both the long and the
short side of the market, one major advantage of the long-short strategy
is the ability to generate “double alpha,” unlike the long-only possibilities
in the mutual fund industry
 Equity Market Neutral: Like the long-short strategy, fund returns are
generated via the exploitation of pricing inefficiencies between
securities. However, equity market neutral strategies also attempt to
limit the overall volatility exposure of the fund by taking offsetting risk
positions on the long and short side, an effort that might also entail
adopting derivative positions. Absent leverage, these strategies are
expected to produce returns of 3%-4% above cash. 2 - 98
 Hedge Fund Strategies (cont.)
 Arbitrage-Based Strategies
 Fixed-Income Arbitrage: Fixed income arbitrage returns are generated via the
exploitation of valuation disparities caused by market events, investor
preferences, shocks to supply or demand, or structural features of the fixed-
income market. Because the valuation disparities are typically small, managers
usually employ leverage to exaggerate returns. The ability to generate alpha is
driven largely by the manager’s skill at modeling, structuring, executing, and
managing fixed-income instruments. In order to extract decent returns, leverage
of 4 to 8x is common.
 Convertible Arbitrage: Convertible arbitrage returns are generated via several
sources, including interest income on the convertible bonds, interest on the
proceeds of related equity short sales, and the price appreciation of the
convertible bonds, as the instruments gradually assume the value of the equity
into which they are exchangeable. The intrinsic value that positions are expected
to converge upon is based on the optionality of the convertibles, a value derived
from the manager’s assumptions about input variables; and is impacted by share
price volatility. Convertible bonds frequently change their character through time
and if the issuer does well, the bond behaves like a stock, if the issuer does
poorly the bond behaves like distressed, and if little happens the convertible will
behave like a bond. Due to the changing characteristics of these securities,
convertibles will usually sell at a discount to their intrinsic value. Leverage is
often employed to enhance returns.

2 - 99
 Hedge Fund Strategies (cont.)

 Arbitrage-Based Strategies (cont.)

 Merger Arbitrage: Merger arbitrage returns are dependent upon
the magnitude of the spread on merger transactions, which are
directly related to the likelihood of the deal not being completed
due to regulatory, financial, or company-specific reasons. As the
probability of the merger improves, the spread narrows,
generating profits for the position.

 Opportunistic Strategies
 High Yield & Distressed: Distressed strategy position returns are
generated if and when the corporate turnaround develops. When
companies are distressed, their securities can be purchased at
deep discounts. As the turnaround materializes, security prices
will approach their intrinsic value, generating profits for the
distressed manager.
2 - 100
 Hedge Fund Strategies (cont.)

 Opportunistic Strategies (cont.)

 Global Macro: Aims to profit from changes in global economies,
typically brought about by shifts in government policy that impact
interest rates, in turn affecting currency, stock, and bond
markets. Participates in all major markets—equities, bonds,
currencies and commodities—though not always at the same
time. Uses leverage and derivatives to accentuate the impact of
market moves. Utilizes hedging, but the leveraged directional
investments tend to make the largest impact on performance.
 Special situations: Special situation returns depend upon a
variety of corporate events. These strategies may involve
restructurings or recapitalizations, spinoffs, or carveouts, and
directional positions that may not be fully arbitraged. Depending
on the manager’s specific strategy, event-driven returns are
realized when the catalyst necessary to release the position’s
intrinsic value takes place.
2 - 101
 Hedge Fund Strategies (cont.)
 Fund of Funds Investing
 Although not formally a separate strategic category, a fund of funds
acts like a mutual fund of hedge funds. The primary benefit to the
investor of a fund of funds position is that is a convenient method for
achieving a well-diversified allocation to the hedge fund investment
space.
 A fund of funds can either offer concentration in a particular strategy
(e.g., long-short equity) and then diversify across different hedge fund
managers—this is a multiple manager approach—or it can diversify
across strategies, which is the multiple strategy approach.
 Another benefit of a fund of funds is that the manager may have access
to some individual hedge fund investments that the investor might not
have otherwise.
 The primary disadvantage to the fund of funds investor is that there will
be an extra layer of fees necessary to compensate the fund of funds
manager. This additional fee can be as high as 3% of the assets under
management.

2 - 102
 Risk and Return in the Hedge Fund Industry
 It is important to note that hedge fund strategies are not riskless. Joseph
Nicholas, author of Investing in Hedge Funds, summarizes the risk-return
tradeoff to the various hedge fund strategies as follows:

2 - 103
 Teacher’s Retirement System of Texas:
Absolute Return Fund Portfolio, July 2004

2 - 104
 Merger Arbitrage Investing: Example
 Suppose the shareholders of Company XYZ
receive an unsolicited cash tender offer for
$30/share. At the time of the offer—which we’ll
assume was a complete surprise—XYZ’s shares
traded for $20.
PT = 30
 Suppose further that shortly after the takeover
announcement—which still must be approved by
regulatory authorities—the price of XYZ’s shares Pm = 28
rise to $28.

 A simple estimate of the market’s implied

probability that the takeover bid will ultimately be
successful is:
[28 – 20] / [30 – 20] = 80%

 Generally, given the pre-announcement price

(Pi), the tender offer (PT) and the post-
announcement market price (Pm) this probability
can be expressed: Pi = 20

[Pm – Pi ] / [PT – Pi ]
2 - 105
 Merger Arbitrage Investing: Example (cont.)

 The essence of merger arbitrage

investing is to try to predict better than
the market which announced deals will
be completed successfully and which
will ultimately fail.

 A merger arbitrage hedge fund

manager will take long positions in
those deals that he or she thinks have
an implied market probability of
success that is too low. Also, short
positions can be taken in those deals
for which the manager’s subjective
probability of success is below that of
the market.

 The chart at the right summarizes a

recent empirical study that looked at
the takeover success probabilities set
by the market for collections of deals
that succeeded (both competitive and
non-competitive tender offers) and a
collection that failed. The display
indicates that the market can, on
average, distinguish good and bad
deals early in the process. 2 - 106
Merger Arbitrage Example: JP Morgan H&Q

2 - 107
Merger Arbitrage Example: JP Morgan H&Q (cont.)

2 - 108
 Comparing Hedge Funds and Mutual Funds
 Investor Characteristics
 Mutual Funds: Mixture of individuals (54%) and institutions (46%);
minimum investments start at $500-$1,000
 Hedge Funds: Some individuals, but mostly institutional, consisting
endowments (58%), corporate pensions (11%), and public pensions
(8%); minimum investments start at $250,000-$1,000,000

 Regulatory Requirements
 Mutual Funds: Highly regulated; investment activities subject to the
Investment Company Act of 1940, also must register with (and be
monitored by) National Association of Securities Dealers and Securities
and Exchange Commission
 Hedge Funds: Except for antifraud standards, they are exempt from
regulation by the SEC under the federal securities laws. Generally not
subject to any limitations in the management of the fund and not
required to disclose information about the hedge fund's holdings and
performance, beyond what the sponsor voluntarily agrees to provide to
investors.

2 - 109
 Comparing Hedge Funds and Mutual Funds (cont.)
 Fees
 Mutual Funds: Both manager fees and sales charges are limited by
federal regulation, which also compels explicit and prompt disclosure of
those fees to investors
 Hedge Funds: Generally, there are no limits on fees. Typically, manager
takes a fee of 1-2% of AUM, plus 20% of the profits over a contractually
negotiated level. Some funds have sales charges as well.

 Leverage/Derivative Use
 Mutual Funds: Investment restrictions often prohibit the use of margin
accounts (91% of all funds) and short sales (69% of funds). Derivative
security prohibitions are less common (about 30% of funds).
 Hedge Funds: Essentially free to follow any investment strategy that is
defined in the contract between investor and manager. In fact, the use
of leverage and short positions are two of the distinguishing features
that separate hedge funds and mutual funds.

2 - 110