Inflation and Hotels.: The of Following
Inflation and Hotels.: The of Following
Inflation and Hotels.: The of Following
66
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
67
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
ment that the industry is now cannot afford mistakes in inflation dollar, or to changes in occupancy
facing. management. Any lessons that can rates-in addition to changes in
Hard times. Even without be learned from past experience the rate of inflation.
rising inflation (much of which can and by challenging existing norms
be attributed to increasing energy should be carefully considered. Setting IZoom-Rates During
cost, which by itself affects the Inflationary Periods
industry in a negative way), the Methodology One must keep in mind that the
hotel industry faces a combination Our study is, essentially, a look
goal of any hotel company is to
of problems that are perhaps worse back in time. To arrive at our raise net profits, not prices. As
now than ever before. conclusions we analyzed data from basic economics (and common
Consider these circumstances: the 15-year period 1975-1989.2 To sense) tells us, raising prices does
overbuilding and excess capacity, identify meaningful relationships not always result in higher profits
cut-throat competition, a drop-off between variables in assessing the or, for that matter, even in higher
in both business and leisure travel, inflation-related performance of revenues. The impact of price
declining occupancy rates, reduced the industry during this time, we changes on profitability depends
profitability, severe short-term applied several statistical on, among other things, price
liquidity problems for many hotels techniques, including correlation elasticity, income elasticity, and
(including some of the largest analysis, multiple regression, and the substitution effect. It also
national companies), and increased stepwise and polynomial depends on supply variables such
numbers of business failures and regressions. For verification of our as the relative competitive position
bankruptcies. All of those situa- results we relied on both linear and of the company.
tions are prevalent in the industry non-linear models, as well as For instance, when price elastic-
today. parametric and non-parametric ity is high, price-sensitive consum-
In addition, those problems are techniques (such as Spearman ers will buy less when prices
coupled with the hotelier’s need to Rank Correlation analysis). increase and more when prices
cope with more-demanding finan- decline. Consequently, a room-rate
cial-institution owners, many of Room Rates and
increase may make a negative
whom are themselves in financial Inflation
incremental contribution to rev-
trouble. As a result, hotel compa- During the period under study, enues and earnings. Similarly,
nies must deal with reluctant room rates raised at a pace
were when income elasticity is high,
lenders, more expensive equity exceeding inflation. In fact, for the consumers with declining incomes
capital due to stock-market price entire post-war period (1946- will buy less.
declines, tax laws less favorable 1989), while the consumer price Consumers also consider the
than in the past, and an apparent index went up by 6.4 times, the prices of substitutes. In our case,
drying-up of foreign investment average room rate increased by a for example, consumers might
funds. factor of 11.9 (see Exhibit 1). consider the cost of domestic versus
Such problems are reflected in However, this well-known foreign travel. This choice is
the stock market’s response to the observation does not necessarily
predominantly affected by the
hotel industry: hotel-industry stock substantiate a direct relation- exchange rate of the U.S. dollar.
prices declined by almost 68 ship between inflation and room- When the dollar is strong (other
percent in 1990.1 This across-the- rate changes because other factors
things being equal), demand for
board price decline for virtually all may have been active at the same hotel rooms in the U. S. suffers as
hotel companies was the steepest time. American hotels become relatively
12-month downturn for hotels since For example, it is possible that more expensive when compared
the Great Depression, and steeper in certain periods room-rate with hotels abroad.
than any other industry group in increases were related to rises in The impact of price changes on
the U.S. during 1990. disposable income, to fluctuations profitability also depends on such
Unfortunately, given such a in the exchange rate of the U.S.
supply variables as competition in
weak starting point, the industry The sources for the macro-economic data
2
the marketplace: when declines in
used in our study include the Board of Governors the producer’s capacity utilization
For a gloomy analysis of the current state of
1 of the Federal Reserve System, the Department
the hotel industry, see: Jonathan Dahl and of Commerce, the Bureau of Economic Analysis,
is evident (such as in the case of
James Carlton, "Rooms to Spare: Hotel Industry the Department of Labor, and the Bureau of hotel overbuilding), other things
Suffers from Glut of Capacity and Faces Big Labor Statistics. Our hotel-specific data were
, November 21,
Deficits," The Wall Street Journal obtained from Pannell Kerr Forster, Laventhol & being equal, prices should be
1990, pp. Al, A6. Horwath, and the Compustat data base. reduced to maintain market share.
68
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
What About Those Other
Factors?
To control for the possible effect of
factors other than inflation, we
applied a multiple-regression
model in which changes in room
rates were chosen as the dependent
variable and several other relevant
variables, in addition to inflation,
were considered as independent
variables. Using this procedure
allowed us to estimate the
contribution of each variable to the
changes in room rates. The results
of this multiple regression, which
are presented in Exhibit 2, point to
the following conclusions.
(1) Inflation was clearly the
dominant consideration in pricing
decisions. The main factor affecting
room-rate changes was a change in
the general level of inflation. As
Exhibit 2 shows, the variable
This exhibit shows that, from 1945-1990, average hotel occupancy, measured
as a percentage of rooms available, has declined while the consumer price
&dquo;Consumer Price Index&dquo; out-
index (i.e., the rate of inflation) has steadily increased. During that same period, weighed all others. The inflation
the average daily room rate also increased except for a slight tailing off in the variable shows both the highest
late 1980s. level of statistical significance and
The raw data show that, for most sub-periods from 1946-1983, room prices the highest coefficient among all
increased at a rate greaterthan the rate of inflation, with the greatest disparity
the factors analyzed. Other critical
occurring during the period 1975-1984. From 1984-1989, such over-
adjustment of room rates (that is, over and above the rate of inflation) was not variables that should have been
as great and, in 1988 and 1989, the trend was actually reversed and room included in the pricing decisions
prices changed at a rate that was less than the rate of inflation. were either ignored or affected the
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
(2) Apparently, income elasticity
(i.e., disposable income) was com-
pletely ignored in the industry’s
pricing decisions. Instead of a
positive relationship between
disposable income and average room
rate, as should have been the case,
there is a negative relationship
between the two. Indeed, the
evidence shows that fluctuations in
disposable income over time affected
pricing decisions in the wrong
direction! The negative coefficient of
1.32 for changes in disposable
income indicates that during this
period the average daily rate went
up by more than 1 percent for every
1-percent decline in disposable
income.
This finding suggests that, by
following a strict policy of pegging
room prices to inflation, the industry
not only ignored the income effect
but actually priced against it.
(3) Changes in the exchange rate
of the U.S. dollar were either
ignored in the pricing decision or
they were considered incorrectly
(prices should have been raised
when the dollar weakens and vice
versa-other things being equal). ing the magnitude of room-rate quality also occurred in most other
The small, negative coefficient for adjustment relative to the rate of items included in the consumer-
the exchange-rate variable (-.149) inflation. When the impact of the price-index &dquo;basket&dquo; that was used
indicates that the industry was other factors is taken into consider- as a bench mark for comparison
following an ill-advised pricing ation, it becomes apparent that the with the hotel industry for this
policy in this respect. industry over-adjusted room rates period.
(4) The data presented in relative to the general rate of To much lesser magnitude, the
a
Exhibit 2 also show that no consid- inflation. During the period 1975- same conclusion of over-adjustment
eration was given to prevailing 89, after adjusting for other factors, is valid for all sub-periods during
business conditions, such conditions the average daily rate went up by 1946-1983. Interestingly, however,
being detectable by changes in the almost 1.8 percent for every 1- from 1984 to 1989, such over-
occupancy rate. The relationship of percent increase in the rate of adjustment was greatly slowed or
the variable &dquo;Hotel Occupancy inflation (as indicated by the occasionally reversed (see Exhibit 1).
Rates&dquo; to hotel-room prices is inflation slope coefficient, beta, of For example, for the two-year period
statistically insignificant (a T-value 1.76, in Exhibit 2).3 While that 1988-1989, the percentage change
of 1.1). This finding suggests that disparity may, in part, represent in the average daily rate was less
the hotel industry ignored feedback improvements in the &dquo;product&dquo; than the percentage change in the
about business conditions and the being sold (i.e., better lodging consumer price index, which is
competitive status of the industry services), the &dquo;you pay for what you illustrated in Exhibit 3. Such a
(as reflected by changes in the get&dquo; rationale should not be over- reaction to inflation can be explained
occupancy measure) when making stated. Improvements in product by intense competition during this
pricing decisions. period and, as we will discuss later,
-
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
yield management and revenue industry by using standard
management. multiple-regression analysis to
The results of our study indicate identify relationships between
that less than full adjustment of inflation and occupancy rates
room rates to keep up with infla- (occupancy rates are the most
tion is often the right policy. This widely watched performance
key point will be clarified below. measure for the industry). Again,
But first, let’s examine whether in order to account for the potential
the policy of excess upward adjust- effects of factors other than
ment of room rates relative to inflation on hotel occupancy, we
inflation was related to the severity included other relevant variables
of inflation. To find out if there was (i.e., changes in disposable income,
a difference in excess room-rate changes in average daily rates, and
increases between high- and low- the exchange rate of the U. ~.
inflation years, we analyzed a non- dollar, as well as a time variable)
linear polynomial regression in our regression. The results of
model.44 this analysis, presented in Exhibit
The results indicate that the 4, clearly indicate that, in recent
over-adjustment of room rates was years, inflation had a negative
not related to the rate of inflation. effect on hotel occupancy rates.
It seems safe to conclude that, as a For the period 1975-89 as a
rule, the policy of over-adjustment whole, for every I-percent increase
was consistently followed in high- in the rate of inflation, hotel
inflation years as well as during occupancy rates declined by about became less of a factor as the rate
those years characterized by low 1.8 percent. This indicates a of inflation declined.
inflation. significant negative relationship of A sporadic review of hotel
Is more better? The ability of almost two-to-one between infla- companies’ revenues, net earnings,
the hotel industry to over-adjust tion and occupancy rates during returns on equity, returns on
room rates relative to the rate of this period (see Exhibit 4). assets, and other performance
inflation is usually considered to be Interestingly, the coefficients are measures during recent years
an industry strength. In fact, that higher for the period 1975-1989, points to the same conclusion:
dynamic is often mentioned by when some of the highest inflation inflation has a negative impact on
industry observers as an indication periods in U.S history occurred, occupancy rates. Regrettably, we
that the hotel industry is a good than for the period 1960-89 as a could not rigorously test those
hedge against inflation. But, is whole. The data indicate that, from variables and separate the impact
such really the case? Let’s examine 1960 to 1989, the Consumer Price of other factors, such as
the effect of inflation on the hotel Index coefficient was -.89, which overbuilding, due to the lack of
industry and then evaluate the suggests that even during a period consistent data over a period long
industry’s room-rate pricing that includes less-severe-inflation enough to satisfy the requirements
policies.5 years, for every 1-percent change of a proper test, such as regression
in the inflation rate hotel occu- analysis.
Inflation and the Hotel pancy declined by almost 1 percent Pricing policies pose a
Industry (after taking into consideration the problem. Our findings seem to
We began an examination of the impact of the other factors we’ve suggest that, at least for the time
effects of inflation on the hotel discussed). This tends to suggest period being studied, the hotel
that the negative effects of inflation industry followed an automatic,
Polynomial regression allows for a non-
4 are greater when inflation is
linear fit. The best fit that we found for these &dquo;nominal-cost-plus&dquo; approach to
variables was still linear, indicating no higher steeper. pricing. As a result, the industry
marginal propensity to adjust room rates when To check this directly we again raised room rates more than
inflation is higher.
For a related discussion of how inflation
5 used a polynomial regression justified by the prevailing economic
affects the cost of capital for hotel companies,
see: Avner Arbel and Robert H. Woods, "Debt
model that confirmed that the conditions during some of the past
Hitch-Hiking: How Hotels Found Low-Cost negative impact of inflation on inflation periods and at other times
Capital," The Cornell Hotel and Restaurant didn’t raise them enough. Overall,
, 31, No. 3 (November
Administration Quarterly occupancy rates was greatest
1990), pp. 105-110. during high-inflation periods and the hotel industry failed to consider
71
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
several important variables in its spite of the fact that in recent years
inflation pricing policies. it actually faced situations quite
similar to the one described above.
A &dquo;Hypothetical&dquo; Case The evidence shows that the
Given that prevailing economic industry’s pricing decisions were
conditions should be part of any largely based on a single criterion:
Simplistic room-pricing room-rate-pricing decisions, the general rate of inflation. The
models amplify the negative consider how the following scenario resulting room-rate adjustments
of inflation and may could affect hotel companies. Keep apparently were automatic reac-
prevent hotels from benefiting in mind that the hotels’ product is, tions to inflation, and ignored other
the opportunity to charge to a large extent, a non-basic good. key factors, such as changes in
even higher prices. Typically, such products have a consumer disposable income, price
72
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
were significantly related to the
rate of inflation but not to other
factors in the correct manner.
Double-checking the results.
Our empirical testing model is
quite straightforward. Therefore,
an important question is whether it
is possible that more-involved
testing models would come up with
different results. For example, is it
possible that a learning process
took place resulting in some
delayed effect that was not cap-
tured by our one-period analysis?
To answer this question we
lagged variables for several differ-
ent time periods. The results of this
lagging procedure indicate that all
inflation-related price adjustments
occurred within one year with no
residual or spill-over effect to
subsequent years. Evidently, then,
the adjustment of room rates to the
rate of inflation-and practically
only inflation-took place within
one year. Furthermore, we found
the results to be virtually insensi-
tive to model-testing specification.8
Sub-Optimal Pricing Policies
Contrary to many industry
observers’ accepted view, the hotel
industry in recent years has not impact of inflation by ignoring both years has been endorsed by the
proven to be inflation proof. The important demand-side and American Hotel & Motel Associa-
fact that room rates were adjusted supply-side variables that should tion (see box on this page). The
to keep pace with inflation or, be included in any rational pricing Hubbart Formula is considered
actually, over and above the decision.Furthermore, this reli- worldwide as an industry standard
general level of inflation, did not ance on simplistic pricing policies and a &dquo;text-book&dquo; approach to
protect the industry from the evils might have prevented the industry hotel-room pricing.
of inflation or from other from benefiting from the opportu- The Hubbart Formula and other
unfavorable developments that nity to charge even higher prices at common pricing methods in the
accompany inflation. some points in time. Overall, the industry are cost oriented. With
On the contrary, the simplistic hotel industry followed a sub- some variations, they all use cost-
pricing model used by the hotel optimal pricing policy. plus approaches while largely
industry amplified the negative How can this rtatapen? One ignoring demand-side and some
should not be surprised by these critical supply-side variables such
8
Other regression models used to test the
results included: log-log, semi-log, polynomial, results, given the archaic and as the competitive position of the
and nonparametric Spearman Rank Correlation.
While the resulting values of the parameters are
fundamentally flawed pricing company resulting, for instance,
methods widely used by the hotel from overbuilding in the area.
slightly different for different tests, each
variable tested showed the same sign (positive or industry. For years the industry These traditional and ineffective
negative) and the same relative importance. has followed a simplistic nominal-
Overall, the tests’ results pointed to the same pricing approaches, which we
conclusions as those discussed above. When the cost-plus approach to pricing. This believe are the wrong tools to use,
variables were lagged to accommodate a possible
is substantiated by the wide use of are reflected in our findings by the
delayed effect, the results were statistically
insignificant. the Hubbart Formula, which for following characteristics.
73
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
-
We indeed found that, as cost- (because one anticipates selling income, a dollar, a decline in
weak
plus approaches call for, during fewer rooms that must still cover all excess capacity in the market, and a
inflationary periods room prices costs). But if you stop and think unique competitive position for the
were adjusted upward in line with about it, given the typical income specific hotel in question).
the nominal increase in cost. The and price elasticities for hotel
It’s Time for a Change
over-adjustment observed in our services, the correct pricing ap-
findings reflects, perhaps, a margin proach is to lower room rates (other Previous studies that identified the
of safety taken by hotels to account things being equal) when consum- hotel industry as a hedge against
for unexpected inflation. ers’ disposable incomes decline or inflation during the 1960s and mid-
o
As cost-plus approaches competition escalates. 1970s clearly indicate much better
require, the upward price adjust- o
As basic economic theory performance than what we found in
ments were implemented as soon as suggests, a simplistic cost-oriented this study for more recent years.9
the cost increases were detected. pricing approach may result in The differences in results highlight
This observation is reflected in our prices that do not clear the market what can go wrong with a simple
regression-analysis results of no lag (the result of which will be a large cost-plus approach to inflation price
effect in price adjustment. inventory of unsold rooms). By adjustments.
®
When using a cost-plus pricing ignoring both supply-side elements While a cost-plus approach may
strategy, &dquo;normal&dquo; profit is treated (relating to competition) and de- be effective in certain kinds of
as a cost item and is automatically mand-side elements, and by not business situations, it does not
included in the price-adjustment taking into consideration the work in all situations. Cost-plus
procedure regardless of whether simultaneous interaction between pricing introduces less bias when
such profit is feasible for the period supply and demand factors, the (1) inflation rates are relatively
under consideration. This is consis- wrong prices may be generated. moderate, (2) inflation is not
tent with our observation of &dquo;no The overall result is low occu- accompanied by declines in eco-
relationship&dquo; or &dquo;wrong relation- pancy rates in the hotel industry nomic activity and disposable
ship&dquo; between changes in room over the years. Furthermore, the income, (3) the level of excess
rates and market conditions. outcome of such policies during capacity as a result of overbuilding
a
According to the Hubbart abnormal periods can be devastat- is not significant and competition is
Formula and other cost-oriented ing. For instance, during periods of not otherwise particularly intense,
pricing methods, an assumption re- high inflation accompanied by and (4) the role of other factors
garding the expected future occu- declines in economic activity and (such as the exchange rate of the
pancy rate must be made in order to income (i.e., during stagflation dollar) is not critical.
calculate the room rates that will periods or when there’s excess Not the time to KISS. Evi-
cover all costs, including return to capacity), and when the exchange dently, the influence of factors other
capital (refer again to the box on rate of the U.S. dollar fluctuates than inflation on the hotel
page 73). Admittedly, this assump- significantly, a nominal-cost-plus industry’s business environment has
tion can be revised over time to suit approach can be painfully sub- grown considerably in recent years.
prevailing business conditions. optimal, as our findings suggest. Unfortunately, in recent inflation
However, when the cost-plus Missed opportunities. Evi- years those factors often acted in
approaches are strictly followed, dently, the industry was caught in a combination for some duration of
such revisions of predicted occu- cost-trap during much of the time time that was long enough to have
pancy rates result in the wrong- covered by our study. In its attempt an effect on overall business results.
that is, inverse-price adjustments. to cover all costs during periods of This suggests that the more in-
Consider what happens when unfavorable market conditions by volved the business situation, the
occupancy tails off, such as during raising room rates, the hotel indus- less appropriate is the simplistic
periods of expected stagflation and try ended up losing sales and cost-plus approach. The common
increased competition. Using the damaging profitability. During other acronym that reminds managers to
Hubbart Formula, one simply re- periods the industry missed the
duces the predicted occupancy rate opportunity to charge higher prices Hotel9 See: Avner Arbel and Paul Strebel, "The
Industry Hedge Against Inflation: The
as a
to show that room sales are ex- when market conditions would have Empirical Evidence," The Cornell Hotel and
Restaurant Administration Quarterly
, 20, No. 3
pected to decline. Using the for- allowed for such increases (e.g.,
(November 1979), pp. 4-7; and Avner Arbel and
mula, the effect on the room rates of those periods characterized by any Paul Strebel, "A Micro Test of the Rational
reducing the predicted occupancy combination of the following: low Expectations Hypothesis: The Case of the Hotel
Industry," Journal of Economics and Business
,
rate is always higher room rates inflation, an increase in disposable 33, No. 1 (Fall 1980), pp. 66-71.
74
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
&dquo;Keep It Simple, Stupid&dquo; (KISS)
doesn’t seem to apply in the case of
room-pricing decisions.
Hubbart the horrible. Given
the widespread use of cost-oriented
pricing methods by the hotel
industry, any results different from
the ones reported in our study
would have been surprising. In
fact, our results indicate that the
Hubbart Formula and its spin-offs
may be the worst thing to happen
to the hotel industry in recent
years.
On the positive side, however,
our data suggest an improvement
in inflation pricing in the most
recent period (1983 to 1989) and an
even greater improvement in the
last two years-1988 and 1989-of
our study period. In fact, during
the two-year period, the room-rate
inflation ratio was significantly
below 1 for the first time.
When considered with other
indicators not included in the
regression analysis presented here,
there is evidence that many hotels
have abandoned the mechanical
approach of adjusting room rates to
match the rate of inflation and that
they now employ more appropriate the industry to use less-simplistic Instead of a mechanical ap-
approaches to pricing, such as yield pricing techniques.&dquo; The results proach to pricing, the hotel indus-
management or the even-better are reflected in our data. try must learn to solve the mystery
approach of revenue manage- of accurate pricing by challenging
ment.10 Consequently, these hotel Message for the Future the existing norms, using all
companies saved themselves from The message to hoteliers regarding relevant economic indicators, and
the cost trap. future inflation periods is clear: looking beyond the simple ap-
The trend toward more-effective using nominal-cost-plus proach for the optimal solu-
pricing policies may be a positive approaches to determine inflation- tions.
outcome of increased competition related room-rate increases is Pricing techniques based on
and the other prevailing business inadequate for the hotel industry. yield management and revenue
conditions discussed earlier. As the lessons of the past indicate, management provide a fresh
Apparently, the recent negative such approaches can be damaging approach because they take into
economic developments have forced because they fail to take into account in a systematic way the
account factors that, depending on combined effect of all relevant
10
The distinction between "yield manage- market conditions, may factors that should be considered
ment" and "revenue management" is important.
While yield management attempts to manage significantly affect the bottom line. in the pricing decision.
the inventory of rooms over time by opening and
closing rates to maximize yield, revenue
Admittedly, such methods are
management takes pricing policy a step further
11
One such new approach is "Market not as easy to implement as cost-
to include principles of optimal price discrimina- Segment Profit Analysis." For a discussion of
tion and simultaneous supply-demand pricing. that strategy, see: Kate D. Dunn and David E. plus approaches. But the hotel
Revenue management, then, is designed to Brooks, "Profit Analysis: Beyond Yield industry can no longer afford to
produce multi-price systems that attempt to Management," The Cornell Hotel and Restaurant make the same pricing mistakes it
clear the market at all times, leaving no, or only Administration Quarterly
, 31, No. 3 (November
a little, excess capacity. 1990), pp. 80-90. has made in the past. 0
75
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015
Downloaded from cqx.sagepub.com at Bobst Library, New York University on May 18, 2015