| The
Benefits of Hybrid Valuation Models
By
David S. Jenkins
JANUARY 2006 - It
is widely accepted that business value is derived from the
stream of discounted future benefits to the owners. The
primary difference among the various valuation approaches
is attributable to the method in which those benefits are
estimated. Controversy exists among valuation practitioners
and academics as to which methods are most appropriate,
as evidenced by, among other things, the substantial amount
of litigation and other legal proceedings surrounding valuation
issues. The importance of developing reliable valuation
estimates cannot be overstated, especially considering their
impact on matters such as estate tax liability, business
acquisitions, buy–sell agreements, and the division
of assets in marital dissolutions. Research on the theory
and application of valuation methods often yields mixed
results, which only exacerbates the debate.
Business
valuation theory recognizes three broad approaches to estimating
value: the income approach, the asset-based approach, and
the market approach. The income approach uses an entity’s
estimated future income stream as a basis for value. The
asset-based approach focuses on determining an entity’s
collective asset values. The market approach can be thought
of as a derivative of the other two approaches (the application
of some market multiple of assets or income). Because the
income and asset-based approaches each demonstrate comparative
advantages and weaknesses, it follows that some form of
hybrid technique that utilizes a combination of income and
asset-based data may be generally superior as a valuation
model.
Income
Approach
The
most common methods of estimating value have traditionally
involved the discounting or capitalizing of an income stream.
In the income approach, variables such as earnings or cash
flows are utilized as a proxy for the expected benefits
to the owners of the business. Common examples of valuation
methods under the income approach are the earnings capitalization
model and the discounted cash flows model.
The
effective implementation of any model with an income approach
requires a reasonable estimate of the expected future benefit
stream as well as an appropriate rate at which to discount
the benefit stream. In a discounting model, a projection
of income is estimated for a finite period, followed by
a terminal value calculation that assumes a constant income
growth rate from that point into perpetuity. The income
stream is converted to present value by applying an appropriate
discount rate. In a capitalization model, a representative
level of income is capitalized into perpetuity at a capitalization
rate determined by the difference between the appropriate
discount rate and a constant, sustainable level of growth
(i.e., a price-to-earnings multiple). The primary difference
between discounting and capitalizing is the level of discretion
afforded in controlling the growth of the income stream.
In any case, reasonable growth assumptions and an appropriate
discount rate are imperative for effective valuation.
Asset-based
Approach
Asset-based
methods typically involve restating both assets and liabilities
to their current values to arrive at a net asset value.
The restatement can be done on an individual component level
(discrete valuation) or collectively (collective valuation).
Given the relative difficulty of individually valuing a
variety of assets, such as real estate, machinery and equipment,
and inventory, it is often necessary to employ valuation
specialists. Collective valuation requires a single analysis,
which identifies the collective value of the assets and
liabilities over and above their recorded GAAP value (i.e.,
a price-to-book multiple). Even with asset-based models,
value remains a function of expected benefits to the owners.
The value of assets is generally derived from either future
income-generating potential or liquidation value, depending
on the circumstances at a given time.
Hybrid
Models
The
income and asset-based approaches to valuation have relative
strengths as well as obvious limitations. For example, the
income approach allows for specific and direct estimation
of future benefits to the owners, which is consistent with
the theory of value. On the other hand, if the estimation
of future benefits is directly based on historical income,
the precision of the estimate will depend heavily on the
persistence embodied in the historical income measure and
on the growth assumptions incorporated into the model. If,
for example, current or historical income contains large
transitory components, the relationship between historical
and future income may be distorted. In addition, to the
extent an inappropriate discount rate is utilized, value
estimates will be adversely affected.
Asset-based
valuation approaches can be effective in that the accurate
identification of individual asset and liability values
will yield a reliable value estimate. In addition, unlike
the income approach, an equity discount rate, the estimation
of which can have a significant impact on the valuation
conclusion, is not required for an asset-based approach.
On the other hand, it is often difficult to accurately restate
GAAP book value to current value for an array of assets,
especially when a significant amount of unrecorded intangible
assets exists.
Implicit
in comparing and evaluating the two approaches is the idea
that each is partly a function of the particular context
in which a company is being valued. A valuation model that
simultaneously utilizes the relative strengths of each approach
may be more effective.
Example
The
following example demonstrates the advantage of a hybrid
approach that incorporates both income and asset data.
XYZ
Company is a supplier of medical monitoring equipment. The
results from the year-end financials show XYZ’s earnings
and book value are $2.4 million and $76.3 million, respectively.
The industry’s comparable price-to-earnings (P/E)
and price-to-book (P/B) ratios are approximately 10 and
1.5, respectively. The resulting value estimate from the
income approach using market comparables to develop the
earnings multiple is $24 million ($2.4 million x 10). The
value estimate from the asset-based approach using market
comparables to develop the book value multiple is $114.5
million ($76.3 million x 1.5). Given that XYZ is a publicly
traded company with a market value of $64.5 million, the
income approach resulted in a significant undervaluation
of XYZ of $40.5 million (or 63% of actual value), while
the asset-based approach resulted in an overvaluation of
$50 million (78%).
A quick
analysis of the information above reveals that XYZ’s
profitability is quite depressed, as demonstrated by a return
on equity (ROE) of approximately 3.7%, as compared to an
industry average ROE of 9%. Despite the weak earnings performance
of XYZ, the market is assigning an earnings multiple of
approximately 27 (64.5 ÷ 2.4), significantly higher
than the industry multiple of 10. Further analysis demonstrates
that the P/B multiple for XYZ is 0.85, quite low compared
to the industry average of 1.5. Because the P/B method is
designed to capture collective asset values above recorded
book value and the market is valuing the company at approximately
its book value, there appears to be no goodwill value. XYZ
has been assigned a large premium on its earnings multiple,
while receiving a significant discount on its book value
multiple.
It
is difficult to reconcile these results by only analyzing
each model individually. Taken collectively, however, income
and asset-based valuations generally yield better valuation
accuracy and more-effective analysis, which is the real
benefit of a hybrid approach. Take, for example, a valuation
model that simultaneously generates earnings and book value
multiples from market comparables, such as a multiple regression
of price on both earnings and book value against a sample
of industry peers. Incidentally, such regressions have been
shown to produce viable estimates of the necessary capitalization
rates for implementation of the excess earnings model, a
popular, albeit controversial, hybrid model. This regression,
performed on a large sample of market comparables from XYZ’s
industry peers, produced earnings and book value coefficients
of 6.843 and 0.556, respectively, which resulted in excess
earnings capitalization rates of approximately 6.5% and
14.6% for tangible and intangible assets, respectively.
Given the above data for XYZ, the value estimate derived
from the earnings and book value coefficients directly is
$58.8 million [($2.4 million x 6.843) + ($76.3 million x
0.556)].
Utilizing
the excess earnings method to estimate the value for XYZ
will, by definition, produce the same valuation estimate
as derived above. In the excess earnings model, value is
calculated as the sum of tangible net assets (in this case
proxied by book value) and capitalized excess earnings.
Excess earnings are calculated as earnings minus net tangible
assets, multiplied by the expected return on net tangible
assets, and then capitalized at the intangible rate to arrive
at an estimate of intangible value, or goodwill. For XYZ,
book value is $76.3 million and excess earnings are –$2.56
million [($2.4 million – $76.3) x 6.5%]. Unsurprisingly,
excess earnings are negative, given XYZ’s relatively
depressed profitability, and as a result, the excess earnings
method yields a value estimate lower than book value, because
estimated intangible value is –$17.5 million (–$2.56
million ÷ 14.6%). The value estimate produced by
the hybrid excess earnings model (which equals the value
estimate derived from the earnings and book value coefficients)
of $58.8 million ($76.3 million – $17.5 million) produced
an undervaluation of only $5.7 million (9%).
The
above results demonstrate that a hybrid approach provides
a more complete analysis of value than either income or
asset data alone. (For a comparison of value estimates of
the hybrid, income, and asset-based approaches, see the
Exhibit.)
First, as previously mentioned, XYZ’s market value
approximates its book value. As shown by the excess earnings
method, this directly implies that XYZ has no positive intangible
value, which can be interpreted in at least one of two ways.
In
the first interpretation, the market thinks that XYZ will
not survive; poor earnings performance, poor cash position,
and large amounts of debt on the balance sheet may be responsible
for this belief. In such an interpretation, XYZ’s
liquidation value is approximated by its book (or tangible
net) value. To the extent that company failure and liquidation
is unlikely, a second possible interpretation is that XYZ’s
book value serves as a better proxy for expected future
normal earnings capacity than do current earnings. As previously
discussed, XYZ’s ROE is low compared to industry standards.
Because research on ROE has shown that profitability (especially
extremely low or high levels) tends to move toward a normal
or average level over time, one would expect XYZ’s
earnings—assuming the company survives or even thrives
in the future—to increase to a level more commensurate
with its book value over time.
Effectiveness
of the Hybrid Approach
Utilizing
a hybrid approach provides for a more comprehensive analysis
of value based on earnings and asset data considered collectively.
More broadly, research reveals that hybrid models are superior
in valuation accuracy to single-variable models in general
as well as in other economic contexts, such as when a company
has significant intangible value. In the final analysis,
the effectiveness of hybrid models is derived from their
ability to simultaneously capture the multiple dimensions
of valuation information contained in income and asset data.
David
Jenkins, PhD, CPA, is an assistant professor of accounting
in the department of accounting and management information
systems at the Alfred Lerner College of Business and Economics
at the University of Delaware, Newark, Del.
|