Summary-Forecasting

1
Summary-Forecasting

• Choose key drivers of performance
• Krispy Kreme-New stores, EBIT
• Begin with a benchmark value-
• Krispy Kreme- End-of-year number
• Then adjust based on additional input
• previous years growth, mgts declared strategy,
  analyst forecasts, etc.

2
Forecasting Using Quantitative Models Based on
Simple Time Series Assumptions

• Quantitative Models can be used to simplify the
  forecasting task
• Example Krispy Kreme Sales Growth
• -Random Walk with a drift
• -Autoregressive with a drift
• -Simple Moving averages
• -Quarterly modeling

3
Valuation- Theory and Concepts
4
Valuation

• The process of converting a forecast into an
  estimate of the value of the firm.

5
Valuation

• Applications
• -capital budgeting
• -strategic planning
• -investing, including mergers and acquisitions
• -credit analysis
• -legal issues, e.g., estate and tax assessment

6
Valuation Approaches

• Discounted cash flow methods
• Discounted abnormal earnings
• Price multiples -- most common practice
• Price to earnings
• Price to book
• Price to sales

7
Valuation Theory

• Three key theoretical questions
• The object What is being valued?
• The context What forces and circumstances drive
  the value of the object?
• The risk How likely are they to occur?

8
The Object

• Equity interest in a firm can be viewed as the
  right to a series of residual future cash flows.
• The cash flows arise from two sources
• Profit generation
• Liquidation
• These cash flows have an expected value.

9
The Context

• What drives the investment value of these cash
  flows?
• Portfolio view- systematic risk
• Traditional view- expected profitability, risk .

10
The Risk

• Some investment risk factors driving the
  valuation of expected future cash flows
• Inflation/deflation
• Economy (e.g., recession vs expansion)
• Risk of credit default and corporate failure
• Competition
• Technological change
• Changing consumer demands
• Changes in the regulatory environment

11
Summary- Valuation Theory

• If markets are at all rational, valuation must
  ultimately be driven by expectations of future
  performance, including earnings and risk.
• Performance and risk are both a function of the
  business economics of a firm.

12
Valuation- The Process

• Investor expectations are formed based on
  reported operating performance to date.
• Financial and accounting analysis bring to light
  current and past operating performance.
• Forecasting brings past and current information
  to bear on the question of what expected future
  performance is.
• Valuation modeling involves the conversion of
  those forecasts into a quantitative assessment of
  value.

13
The Theoretical Basis for All Traditional
Valuation Models

• The dividend discount model

14
Valuation by Formulae

• Gordon-Shapiro growth formula
• P d/(r-g)

15
Gordon-Shapiro

• Assume a constant growth rate in dividends
• Let g the dividend growth rate

16
Gordon-Shapiro

• Multiply both sides by (1r)/(1g)

17
Gordon-Shapiro

• Assume that dividends are paid out of current
  earnings, E, with a payout ratio k, and get

18
The Problem with Dividends

• Dividend policies are sticky, i.e., often slow to
  reflect economic change.
• Dividend models usually require predictions of
  huge liquidation payouts at some end point.
• Some companies pay no dividends at all.

19
Concepts

• Valuation requires the input of numerical
  forecasts of expected future cash flows. These
  numerical inputs can be categorized as
• Cash basis, or
• Accrual-based.

20

• Cash Basis gt
• The Discounted Cash Flow Model
• Accrual Basis gt
• The discounted abnormal earnings model
• Price multiples (PE, Price to book)
• Discounted ROE model

21
If cash flows are the object, why bother with
accrual numbers?

• I am always amazed, in a sense, that we spend all
  of this money in producing accounting numbers.
• Finance people tell us that that we should spend
  resources to undo what we spent all of that money
  to do in the first place in order to find value.
  (A little short blurb comes out about every
  month in Business Week, Forbes or some other
  magazine telling us that we must find cash)

22
Why accounting analysis as oppose to cash flow
analysis

• focuses attention on value drivers that have been
  identified previously as key to measuring
  performance e.g., ROE
• Less apparent estimation of the distant future.
• To the extent the accrual system maps well to
  underlying economic phenomena, can be easier to
  forecast than raw cash flows.

23
What does some of the research say?

• Numerous studies starting with Ball and Brown
  (JAR, 1968), have shown that stock prices react
  to accounting earnings announcements.
• A whole research industry has been born around a
  thing called earnings response coefficients (ERC)

24
What does some of the research say?

• Easton, Harris, and Ohlson (JAR, 1992) found that
  over long event windows that earnings account for
  over 60 of the variation in stock returns.
• Dechow (JAE, 1994) recently, in testing competing
  models, found accruals were more closely related
  to stock prices than were cash flows

25
Abnormal Earnings models

• Do you first have to convert accounting numbers
  to cash flow and then discount the cash flows as
  finance theory says or
• Can the accounting numbers themselves be
  discounted?
• Ohlson contends he latter is better

26
The Ohlson Model
27
Ohlson Model

• Assumes clean surplus accounting
• bvt bvt-1 xt - dt

28
Bernards Test of Ohlson Model
29
Bernards Test

• Regressed price over Value Lines earnings
  forecast and current book value
• Results
• intercept bv t1 t2 t4
  R2
• 5.82 1.04 3.18 1.58 6.15 .68
  
• (3.85) (5.40) (3.25) (1.26) (2.83)
• R2 of .29 using dividends

30
What about accounting methods

• We have been examining firms accounting policies
• Ohlson tells us that accounting choices are self
  correcting
• For example, accelerated depreciation writes off
  more of the asset in the earlier years but
  ultimately the same purchase price is written off

31
If Valuation Models are unaffected, why do firms
think accounting choices matter?

• Accounting choices reflect management thinking.
  For example a switch to st. line depreciation
  could signal that earnings are on a real
  decline. A switch to LIFO could signal the same
  thing, but it could also signal that management
  is seeking to save taxes..

32
If Valuation Models are unaffected, why do firms
think accounting choices matter?

• Accounting choices have an impact on the amount
  of value captured between the short and long run.
  High quality permits a more complete reflection
  on value in the near term.

33
If Valuation Models are unaffected, why do firms
think accounting choices matter?

• If analysts are unaware of accounting choices,
  and thus dont see the impact they have, and will
  have, on reported profitability, then firm value
  can be over- (under) stated.

34
Questions to Ponder

• What is the impact of conservative accounting
  Choices?
• What is the impact of liberal accounting choices?
• What about unbiased accounting choices?
• Theoretically, in a perfect world (i.e., free
  markets, lots of competition, and complete,
  unbiased reporting), ROEs approach the cost of
  capital

35
Linkage to drivers

• ROE is familiar
• Can fiddle around with the Ohlson model and
  derive the expression found on page 7-7
• The ratio of valuation to book becomes discounted
  future abnormal ROE

36
Linkage to drivers
37
A Related Model That Focuses on Operating Earnings

• Debt Equity / Book
• 1 ?(NROAi - WACC) / (1WACC)i ,
• Where
• NROA return on net operating assets, and
• WACC weighted average cost of capital, i.e.,
  the discount rate for the business operation
  itself.

38
Shortcut Methods-Example
39
Impact of accounting methods on terminal values

• In the accounting model the terminal value
  includes only the abnormal earnings beyond the
  terminal year
• Present value of normal earnings is reflected in
  book value
• In the presence of perfect accounting, abnormal
  earnings should be 0, long-term.

40
Impact of accounting methods on terminal values

• Approach that is most convenient when accounting
  method choices are unbiased
• Produces an ROE that ultimately equals to the
  cost of capital
• Subsequent Abnormal earnings would be zero and
  terminal value would be nil

41
Abnormal Earnings-Other Modeling Options

• Assumed Random walk
• V BVE AE/r
• Assumed Random walk with decay
• VBVE ?AE / 1-r-B, where
• B rate of decay in abnormal profits

42
Abnormal Earnings-Other Modeling Options

• Assumed future mean reversion, e.g., steady state
  ROE and book equity growth (g)
• P/B 1 (ROE-r) / (r- gBVE),

43
Valuation based on price multiples

• DCF and Accounting Discount methods place heavy
  demands on the person doing the analysis
• One alternative is to let the all seeing, all
  knowing market do it for you by examining the
  price ratios of comparable firms

44
Price Multiples and Comparable Firms

• Comparable firms-
• -same size?
• -same industry?
• Industry averages

45
Price Multiples -- what do you use?

• Trailing PE
• Leading PE
• Unlevered price/sales
• Levered price/sales
• Price-to-book
• Price-to-cash flow

46
Price to earnings ratios

• Vary positively in relation to expected to growth
  in earnings
• Vary negatively with risk
• If no growth in earnings is expected then PE
  should be approximately equal to the reciprocal
  of cost of capital ( For Compaq 1/.13 7.7)

47
Price-to-book

• Should vary according to differences in their
  future ROEs
• Fama and French (JoF, 1992) found that Size and
  Book-to-Market combine to capture the
  cross-sectional variation in average stock
  returns associated with ?, size, leverage,
  book-to-market, and earnings-price ratios. (The
  famous paper that got popularly labeled as beta
  is dead)

48
The Theoretical Relation Between P/E and P/B

• P/B
• If book value equity capital, and
• If price is discounted actual earnings, then
• P/B is the relation between profit earned on
  equity and the cost of equity capital
• If ROECOEC, then P/B should 1
• P/E
• P/B / ROE P/E

49
Price-to-sales ratios

• Should vary with expected profit margins
• Firms with higher expected profit margins should
  be higher in sales

50
Price-to-cash-flow-ratios

• Noisy unless modified
• EBITDA (earnings before interest, tax,
  depreciation and amortization) is frequently used
• Forecast of EBITDA not widely available

51
Valuation by Formulae

• Gordon-Shapiro growth formula
• P d/(r-g)

52
Gordon-Shapiro

• Assume a constant growth rate in dividends
• Let g the dividend growth rate

53
Gordon-Shapiro

• Multiply both sides by (1r)/(1g)

54
Gordon-Shapiro

• Assume that dividends are paid out of current
  earnings, E, with a payout ratio k, and get

55
Miller and Modigliani

• Consider where growth may continue for any number
  of years but is relevant to the firms value for
  only T years

56
Detailed versus Simple

• DCF and discounted abnormal earnings supply the
  proper conceptual framework
• Multiples may be easier and more useful.
• Lets the market decide
• Assumes some circularity

57
Summary

• All traditional valuation models derive from the
  simple dividend discount model.
• Valuation models differ in three key areas
  focus, structure, and terminal value estimation.
• The choice of which model(s) to use and rely on
  depends on the nature of the prediction task and
  the information upon which forecasts are based.