Market valuation and earnings manipulation

1
Market valuation and earnings manipulation

• Shing-yang Hu
• and
• Yueh-hsiang Lin

2
Motivation

• How firms are priced? What determines firm value
  or return (the difference of the value)?
• Market valuation has its impact
• Subrahmanyam and Titman (2001) stock prices can
  affect fundamentals. A higher stock price may
  signal that the firm has a good product, and
  induce consumers to adopt its product.
• Literature of behavior corporate finance
  managers corporate decisions such as investment,
  financing, dividend, MA may be responses of
  market valuation.

3
Motivation

• Scandals in early 2000s Enron, WorldCom,
  eToy,etc.
• A lesson to be learned trying to maintain a
  higher market valuation, managers may take action
  such as earnings manipulation that will reduce
  long-term firm value.
• Paul Volcker (the former Federal Reserve
  Chairman)
• Optimistic visions of a new economic era set the
  stage for an explosion in financial values. It
  was an environment in which incentives for
  business management to keep reported revenues and
  earnings growing to meet expectations were
  amplified.

4
Motivation

• Jensen (2005) agency costs of overvalued equity
• Overvalued EquityManagerial Heroin. It feels
  great at the beginning. But lead to massive pain
  in the end.
• Managers use overvalued equity to make
  acquisitions to satisfy growth expectation.
• Managers use access to cheap debt and equity
  capital to engage in excessive internal spending
  unprofitable risky greenfield investments.
• Finally, they turn to manipulation and even
  fraud to satisfy the markets demands.
• Equity-based incentive compensation cannot solve
  the problem. High equity or option holdings are
  like throwing gasoline on a fire.

5
Objective

• There is no study so far to empirically examine
  the relation between market valuation and future
  earnings manipulation.
• We fill a gap in the literature by providing
  evidence of the relation.
• We ask
• Does the relation exist?
• Why does it exist?
• Can good governance mechanism alleviate this
  problem?

6
Hypotheses

• Manipulation hypothesis Managers of firms with a
  higher valuation will manipulate future earnings
  more (have larger future accruals) to meet
  investors expectations.
• Based on the assumption that keeping higher
  expected growth rates of cash flows needs larger
  reported earnings.
• Expected growth hypothesis Managers of firms
  with a higher valuation expect future performance
  better and increases accruals accordingly.
• For example, managers may build up inventory to
  meet growing demand.

7
Findings

• In our sample (all non-financial companies listed
  on the NYSE, Amex, and Nasdaq in the period of
  1988-2004), averagely, there is a positive
  relation between market valuation and future
  accruals.
• The manipulation hypothesis is more pronounced
  only in firms with limited attention or poorer
  performance.
• For firms with more attention or better
  performance, the evidence does not support the
  manipulation hypothesis.
• Higher valuation firms that issue equities are
  especially aggressive in accounting accruals.
• Can good governance alleviate the incentive to
  manipulate earnings by higher valuation
  companies? Yes.
• No evidence that high valuation companies with a
  stronger governance system or equity-based
  compensations will manipulate more.

8
Methodology

• Measure of market valuation
• Market-to-book ratio of equity
• Price-to-earnings ratio or the price-to-cash flow
  ratio?
• May be inappropriate.
• For example, if earnings and accruals are
  positively correlated and accruals are
  autocorrelated, a higher price to earnings ratio
  will be correlated with a higher future accrual
  due to its definition.

9
Methodology

• Measures of earnings manipulation total accrual
• Calculated from items of the balance sheet (BS)
• Calculated from items of the cash flows
  statements (CF)
• Collins and Hribar (2002) argue that accruals
  from the BS may be biased due to the reasons such
  as MA. Therefore, here we only present results
  on accruals from the CF.
• The data on the CF is only available from 1987,
  and the number of firms reported in 1987 is
  limited so we choose to start the sample from
  1988.

10
Methodology

• To detect the effect of the normal business
  requirements on the total accruals, we use two
  conventional discretionary accrual models
• Jones model JA
• Modified Jones model MJA

11
Manipulation or expected growth hypothesis?

• One-way grouping method For each fiscal year t
  we divide the sample into quartiles based on the
  market-to-book ratio of equity (M/B).
• Test the difference in mean and median of
  accruals for year t1 between the highest and
  lowest M/B quartiles using t and the Wilcoxon
  statistics.
• Under the two hypotheses, the differences should
  be positive.
• If valuation proxies for expected growth, after
  controlling performance, the differences should
  be less pronounced.
• Follow Kothari, Leone, and Wasley (2005) to
  calculate ROA-matched accruals to distinguish the
  two hypotheses.

12
Manipulation or expected growth hypothesis?

• The differences in mean and median of accruals
  between the highest and lowest quartiles are
  significantly positive.
• The differences in ROA-matched accruals between
  the highest and lowest quartiles are still
  significantly positive.
• Consistent with the manipulation hypothesis.

13
A closer look at the Manipulation Hypothesis

• What characteristics of firms will have larger
  differences in ROA-matched accruals between the
  highest and lowest M/B quartiles?
• Manipulation Hypothesis (MH) 1 The difference is
  larger for firms that less attention from
  investment communities.
• When attention is low, the information
  environment will be poor so that investors are
  not able to differentiate a true earning from a
  fake one.
• Two-way independent sorting based on M/B and
  attention, in which we use three measures to
  proxy for attention the inclusion in the SP
  1500, the tracking by the Investor Responsibility
  Research Center (IRRC), and market
  capitalization.

14
Tests of H1

• The differences are significantly positive only
  for less attention firms.
• Results are consistent with MH 1.

15
Manipulation Hypothesis (MH) 2

• Manipulation Hypothesis (MH) 2 The difference is
  larger for firms with poorer operating
  performance.
• A high valuation company face much more pressure
  to improve its operating performance when it is
  poor than it is good.
• Two-way independent sorting based on M/B and ROA.
• Results are consistent with MH 2.

16
Manipulation Hypothesis (MH) 3

• Manipulation Hypothesis (MH) 3 The difference is
  larger for firms that have equity offerings.
• Given the earnings manipulation, the stock price
  will be higher and the SEO will be more
  attractive.
• Two-way independent sorting based on M/B and the
  amount of the sale of equities (SEO).
• Results are consistent with MH 3.

17
Manipulation Hypothesis (MH) 4

• Manipulation Hypothesis (MH) 4 The difference is
  smaller for firms with a good governance system.
• In a good governance system the manipulation is
  more likely to be found out, and managers may be
  replaced.
• Use two measures to proxy for the alignment of
  the governance mechanism
• Governance Index (GI) introduced by Gompers,
  Ishii, and Metrick (2003) from the IRRC database.
• Incentive ratio (IR) used in Bergstresser and
  Philippon (2003) from the CEOs company stock and
  option holdings in the ExecuComp database.
• Regress the discretionary accruals on the
  valuationGI and valuationIR as well as the
  aforementioned variables.

18
Tests of H4

• Negative VGI The better the governance is, the
  smaller the impact of valuation on accruals.
• Negative VIR The stronger equity-based
  compensation is, the smaller the impact of
  valuation on accruals.
• Consistent with MH4 statistically but the
  magnitude is so small that the effect is weak
  economically (compared with coefficients on V).
• Most of the four hypotheses are supported.

19
Conclusions

• We provide the evidence of the relation between
  market valuation and future earnings
  manipulation.
• In the U.S. companies, the relation occurs only
  in firms with limited attention or poorer
  performance.
• For firms with more attention or better
  performance, there is no significant relation.