|
I have a collection of research related material here
that can be downloaded.
Surprising absence of scale for forecast error and forecast dispersion distributions
updated 10/2008.
ABSTRACT
While levels of actual and consensus forecast earnings per share (EPS) vary with scale
(measured typically by share price), magnitudes of the difference (or forecast errors)
do not vary with scale. That is, forecast errors within a certain range
(e.g., +/- 5 cents per share) are equally likely for both high-price and low-price shares.
We also find a similar lack of variation with scale for forecast dispersion,
representing magnitudes of the difference between individual forecasts and the
consensus (mean) for that firm-quarter. The prior literature has assumed that
magnitudes of forecast errors (representing predictability) and forecast dispersion
(representing disagreement across analysts) vary naturally with scale and has
deflated both variables accordingly. Our results suggest that either both variables
do not vary naturally with scale or they vary with scale but other effects reverse
that variation. Our finding that both variables exhibit scale variability around
stock splits is consistent with the latter. In either case, deflation can cause biased estimates.
Revisiting the Basu (1997) estimate of conditional conservatism
updated 10/2008.
ABSTRACT
Basu (1997) offers a measure of ex post accounting conservatism, based on the excess
of the timeliness with which observed bad news is recorded in earnings over that for
good news. Timeliness is represented by the slope of a regression of earnings (scaled
by lagged price) on returns. We provide new evidence which suggests that those
conservatism estimates be used with caution because they mainly reflect the lagged
E/P and loss firm effects, two effects unrelated to conditional conservatism. Observing
similar results when earnings in the timeliness regressions is replaced by lagged earnings
suggests that timeliness is affected substantially by the E/P anomaly (complex relation
between current returns and lagged earnings-to-price ratios), since lagged earnings
should not reflect current period news. The loss firm effect is supported by our finding
that the timeliness coefficient for bad (good) news subsamples is positively (negatively)
related to the fraction of firms reporting losses in those subsamples. We believe that prior
findings based on time-series and cross-sectional variation in the Basu measure are
potentially confounded by corresponding variation in the E/P and loss firm effects. We
illustrate that potential by investigating variation in the measure across book-to-market
ratios and size. Replacing earnings levels in the timeliness regressions with earnings
changes, to control for the lagged E/P effect, reduces substantially estimates of
conservatism as well as the bias due to the loss firm effect.
Tax benefits of LILO transactions
updated 1/2007.
ABSTRACT
Tax shelters have received recent scrutiny in the financial economics literature because
of their impact on firm decisions. While the source of tax benefits for many types of
shelters are fairly straightforward, they are not as evident for Lease in Lease out
(LILO) transactions. This paper uses spreadsheet models to understand how tax benefits
are generated in LILO transactions and to identify factors that enhance the magnitude
of those tax benefits.
Tax expense surprises and future returns
updated 12/2007.
ABSTRACT
We investigate whether surprises in quarterly tax expense predict future returns, after
controlling for surprises in after-tax book income. We find that seasonally-differenced
quarterly tax expense, our proxy for tax expense surprise, is positively related to future
returns over the next two quarters. We confirm that this anomalous link is separate from
other anomalies documented in the prior literature, such as size, book-to-market, accruals,
and price momentum, as well as two anomalies related to tax variables. While higher expense
might intuitively imply bad news, in this case higher tax expense signals good news as it
is positively related to pre and after-tax income. Our results suggest that this good news
is incorporated in stock prices with a delay because investors do not recognize fully the
ability of tax expense surprises to predict two key variables that are released in the next
two quarters—future book income and future tax expense.
Overreaction to intraindustry information transfers?
updated 12/2007.
ABSTRACT
Prior research has documented that earnings announcements provide information not only
about the announcing firm but also about other firms in the same industry. We document
a stock market anomaly associated with this phenomenon of intra-industry information
transfers by showing that the stock price movements of late announcers in response to
earnings reported by early announcers are negatively correlated with the subsequent price
responses of late announcers to their own earnings reports. Apparently, the stock market
overestimates the intra-industry implications of early announcers’ earnings for late
announcers’ earnings, and that overestimation is corrected when late announcers
disclose their earnings.
Inflation illusion and stock prices: Comment
updated 7/2007.
ABSTRACT
The results reported in Campbell and Vuolteenaho (2004) suggest that stock markets suffer
from massive inflation illusion as suggested by Modgliani and Cohn (1979). That is,
investors incorrectly project the same nominal growth during high and low inflation,
when they should project the same real growth. As a result stock prices are too high
(low) when expected inflation is low (high). We reinvestigate their data and find that
their results are sensitive to the sample period studied, the proxy used for expected
inflation, the use of dividends versus earnings yields, and the VAR methodology employed.
We suggest that it is premature to conclude that the stock market suffers from inflation illusion.
Accounting rules and the relation between earnings yields and inflation
updated 8/2007.
ABSTRACT
Financial economists expect assets with real cash flows, such as inventory, plant, and property, to have
earnings yields (ratio of next period’s nominal earnings divided by current price) that are “real”, in the
sense that they should not vary with expected inflation. The evidence, however, suggests that earnings
yields do in fact vary over time with inflation. One explanation suggested for this contrary evidence is
that investors adjust nominal discount rates for expected inflation but not nominal growth rates. I claim
that such a drastic conclusion is premature because the original arguments for real earnings yields do
not incorporate the accounting rules underlying reported earnings. I construct spreadsheet simulations
that project accounting earnings for different asset classes and find that earnings yields for real assets
should in general vary with inflation. The intuition for my findings is straightforward: because inflationary
holding gains are generally included in accounting earnings, higher inflation is associated with higher
nominal earnings, which then implies higher earnings yields.
Don't fight the Fed Model
updated 4/2008.
ABSTRACT
Over the past decade two remarkable regularities have emerged regarding
expected inflation’s impact on earnings yields and growth: a) stocks
in recent years are priced like Government bonds (forward earnings
yields for stocks move with long-term risk-free rates and expected inflation),
and b) analysts’ forecasts of nominal earnings growth remained steady,
around 12 percent, despite substantial variation in expected inflation.
The first finding, known as the Fed model, has raised fears of inflation
illusion; i.e., investors mistakenly project the same nominal growth rates
during high and low inflation periods, when they should project the same
real growth rate for firms holding real assets. Cliff Asness, in his 2003
FAJ paper, urges investors to “fight” the Fed model. The second regularity
confirms that fear. Despite the logic and evidence supporting inflation
illusion, the investor naiveté implied is so large that it strains credulity.
We take the opposite tack and investigate the possibility that these two
findings are consistent with a rational market. We consider the accounting
rules underlying reported earnings to show why the recording of inflationary
holding gains causes earnings yields to move with inflation (the first
regularity). And we show that growth forecasts are confirmed by observed
growth; i.e., it is rational to forecast nominal growth that does not
vary with inflation (the second regularity) because real growth is in
fact negatively related to expected inflation. Overall, readers should
embrace the Fed model because it yields important insights about stock
market valuation.
Tip of the Iceberg?
(Mathur/Kirschenheiter/Thomas thought piece) updated 8/18/02.
Contributed capital versus retained earnings: tax differences and value
implications
(Li/Shackelford/Thomas working paper) updated 9/3/02.
ABSTRACT
The investor-level tax rate on dividends that is capitalized in share
prices is an important issue that remains unresolved. Recent research has
examined whether the tax code feature that exempts dividend taxes on
return of contributed capital can be used to infer the extent of dividend
tax capitalization. Intuitive arguments have been made by Harrris and
Kemsley (1999) to generate predictions for how the coefficients estimated
from regressions of stock prices on book value of equity and trailing
earnings should vary with the proportion of equity represented by
contributed capital. While the observed results suggest that share prices
capitalize dividend taxes at a fairly high rate, some have questioned the
regression specification used, since these predictions are not formally
derived, and also raised concerns about empirical issues relating to the
results. We return to the primitive stream of taxable and tax-exempt
dividends and derive an alternative specification. Our results do not
support the hypothesis that stock prices distinguish between contributed
capital and retained earnings. We are, however, unable to offer evidence
relating to the underlying question regarding the extent of dividend tax
capitalization.
Accounting for employee stock options
(Mathur/Kirschenheiter/Thomas working paper) updated 12/9/03.
ABSTRACT
Accounting for employee stock options is affected by whether outstanding
options are viewed as equity or liabilities. The common perception is that the
FASB’s recommended treatment (per SFAS 123), which is based on the options-as-equity
view, results in representative financial statements. We argue that this treatment
distorts performance measures for three reasons. First, the deferred taxes associated
with nonqualified options should also be included as equity, but are not. Second,
since unexpected share price changes affect optionholders and equityholders differently,
combining their interests provides an average earnings effect that is not representative
for either group. We show that efforts to isolate the interests of common stockholders
via diluted earning per share calculations (per SFAS 128) are inherently incapable of
identifying wealth transfers between stockholders and optionholders. Finally, projections
of future cash flow statements prepared under SFAS 95 overstate cash flows to current
equityholders by the pre-tax value of projected option grants. We show that these
distortions can be avoided simply by accounting for options as liabilities at grant
and thereafter recognizing changes in option values (similar to the accounting
for stock appreciation rights).
Our analysis of stock option accounting leads to two, more general
implications: a) all hybrid securities issued by firms, other than common
equity, should be treated as liabilities, thereby simplifying the equity vs.
liability distinction, and b) these liabilities should be recorded at fair values,
thereby obviating the need to consider earnings dilution.
Another look at P/E ratios
(Thomas/Zhang working paper) updated 2006.
ABSTRACT
Price-earnings (P/E) ratios should be positively related to growth and negatively related
to interest rates and risk. Whereas earlier investigations of the determinants of P/E
ratios find these links to be weak, results of recent research estimating the cost of
capital imply stronger links. The two sets of research employ different sets of proxies
for P/E ratios (trailing vs. forward ratios), growth (observed vs. forecast growth) and risk.
We update the literature investigating the determinants of P/E ratios by contrasting the
results based on the two sets of proxies. We also investigate a recent finding which
suggests that forward P/E ratios are negatively related to the volatility of reported
earnings, even though reported earnings volatility does not appear in the relation derived
for forward P/E ratios. Our results suggest an indirect connection: firms with lower
earnings volatility, due to lower cash flow volatility and greater earnings smoothing
due to accruals, are associated with higher growth prospects and lower risk.
Is Cash flow King in Valuations?
(Liu/Nissim/Thomas 2007 paper in FAJ) updated 2/12/07.
ABSTRACT
Contrary to the common perception that operating cash flows are better than accounting earnings
at explaining equity valuations, recent studies suggest that valuations derived from industry
multiples based on reported earnings are closer to traded prices than those based on reported
operating cash flows. The question addressed in the article is whether the balance tilts in favor of
cash flows when the following are considered: (1) forecasts rather than reported numbers, (2)
dividends rather than operating cash flows, (3) individual industries rather than all industries
combined, and (4) companies in non-U.S. markets. In all cases studied, earnings dominated
operating cash flows and dividends.
On the informational usefulness of R&D capitalization and amortization
(Lev/Nissim/Thomas working paper) updated 3/15/02.
ABSTRACT
Under U.S. GAAP, reported balance sheet and income statements are based on
immediate expensing of R&D expenditures. We capitalize those expenditures
and derive adjusted equity book values and earnings using simple
amortization techniques (straight-line over assumed industry-specific
useful lives). After confirming that such adjustments increase the
association of book values/earnings with contemporaneous stock prices (and
future earnings), we examine the relation between those adjustments and
future returns. Despite the approximate nature of our adjustments, they
predict stock price movements over the next 20 months. Apparently,
capitalization and amortization of R&D provides information not fully
reflected in stock prices.
Inventory changes and future returns
(Thomas/Zhang working paper) updated 12/29/01
ABSTRACT
We find that the negative relation between accruals and future abnormal
returns documented by Sloan (1996) is due mainly to inventory changes. We
propose three explanations for this result, derived from the prior
literature, but find evidence inconsistent with all three explanations. To
assist future investigations in formulating additional explanations, we
document several empirical regularities for extreme inventory change
deciles. We speculate that demand shifts explain our results, and examine
the feasibility of alternative reasons for the stock market's apparent
inability to recognize the impending profitability reversals. Our evidence
is consistent with earnings management masking the implications of demand
shifts.
Equity valuation using multiples
(Liu/Nissim/Thomas, forthcoming in Journal of Accounting Research)
updated 4/10/01.
ABSTRACT
We examine the valuation performance of a comprehensive list of value
drivers and find that multiples derived from forward earnings explain
stock prices remarkably well for most firms: pricing errors are within 15
percent of stock prices for about half of our sample. In terms of relative
performance, the following general rankings are observed consistently each
year: forward earnings measures are followed by historical earnings
measures, cash flow measures and book value of equity are tied for third,
and sales performs the worst. Contrary to the popular view that different
industries have different "best" multiples, we find that these overall
rankings are observed consistently for almost all industries examined.
Adjusting the ratio formulation typically followed in practice to allow
for an intercept offers some improvement, especially for multiples that
perform poorly. No improvement is observed, however, when we consider more
complex measures of intrinsic value based on short-cut residual income
models (where forward earnings are combined with book values, estimated
discount rates, and generic terminal value estimates). Since we require
analysts' earnings and growth forecasts and positive values for all
measures, our results may not be representative of the many firm-years
excluded from our sample.
Equity premia as low as three percent? Evidence from analysts' earnings
forecasts for domestic and international stocks
(Claus/Thomas forthcoming in Journal of Finance) updated 11/14/00.
NOTE
This paper is a combination of the following two papers. Download them
if you want additional details of the analyses in the combined paper.
The equity risk premium is much lower than you think it is:
empirical estimates from a new approach
(Claus/Thomas 1999 working paper) updated 5/18/99.
ABSTRACT
We offer ex ante estimates of the equity risk premium based on forecasted
accounting numbers. Although our approach is isomorphic to dividend
growth models, it generates various diagnostics that help to narrow the
range of reasonable assumed growth rates. Our results, based on IBES
consensus earnings forecasts over the 1985-1998 period, contrast sharply
with those of prior research. Our estimates of risk premium are
considerably lower than (about 3 percent) the estimates commonly cited
(about 8 percent), and are also more stationary over time. This result
has important implications both for academe (e.g., the equity premium
puzzle) as well as practice (e.g., discount rates for valuation and
over-valued stock markets).
Measuring risk premia using earnings forecasts:
an international analysis
(Claus/Thomas 1999 working paper).updated 5/15/99.
ABSTRACT
The returns earned by US stocks since 1926 have generated an "equity
premium puzzle", since they exceed estimates derived from theory, from
other periods and markets, and from surveys of investors. To determine if
this historic estimate is biased upward, we offer a new approach based on
accounting data and analysts' earnings forecasts, which is used to examine
six different equity markets: Canada, France, Germany, Japan, UK, and US.
For each year between 1985 and 1998, our equity premium estimate is the
difference between the implied discount rate (that equates current prices
with the present value of future flows) and the prevailing 10-year
risk-free rate. These ex ante estimates, which are around three percent or
less in the six markets examined, are support the view that the historic
estimate is too high. Although the accounting flows are isomorphic to
dividends, our approach makes better use of information available
currently and generates diagnostics that help to narrow the range of
reasonable growth rate assumptions.
Stock returns and accounting earnings
(Liu/Thomas published in Journal of Accounting Research).updated 7/15/99.
ABSTRACT
Although most market-based accounting research is based on regressions of
abnormal returns on contemporaneous unexpected earnings, many have
despaired about the intrinsic ability of accounting earnings to explain
stock returns. These regressions exhibit low R2, lower than expected
coefficients on unexpected earnings (ERC's), and various unusual features
including a) non-linearity, b) lower R2 and response coefficients for loss
firms, and c) lower R2 and response coefficients for high-growth and
high-tech firms. Some improvement in explanatory power has been achieved
by including various proxies for information that is currently available
about future period earnings. This paper contributes to that line of
research by deriving a specification, from the abnormal earnings model,
that extends the traditional ERC regression by including current period
forecast revisions of future period earnings (up to year +5). Relative to
the traditional regression, the full specification increases R2
substantially, reduces the bias in coefficient estimates (caused by
omitted correlated variables), and mutes the three unusual features
mentioned above. Fundamentally, we find that revisions in near term
earnings forecasts and discount rate changes are the two most important
drivers of returns. High R2 values can be achieved by simple
specifications that combine these two items (e.g., capitalized earnings,
or the ratio of near term forecasted earnings to the discount rate).
However, the coefficient estimates are less stationary and deviate more
from their predicted values, relative to those for the more complex
specification we derive.
Potential errors in detection of earnings management: reexamining
the studies of the AMT of 1986,
(Choi/Gramlich/Thomas, forthcoming in Contemporary Accounting
Research).updated 8/22/01.
ABSTRACT
We seek to document errors that could affect studies of earnings
management. The book income adjustment (BIA) of the alternative minimum
tax (AMT) created apparently strong incentives to manage book income
downward in 1987. Five earlier papers using different methodologies and
samples all conclude that earnings were reduced in response to the BIA.
This consensus of findings offers an opportunity to investigate our
speculation that methodological biases are more likely when there appear
to be clear incentives for earnings management. A reexamination of these
studies uncovers potential biases related to a variety of factors,
including choices of scaling variables, selection of affected and control
samples, and measurement error in estimated discretionary accruals. And a
reexamination of the argument underlying these studies suggests that the
incentives to manage earnings are less powerful than initially predicted,
and are partially mitigated by tax and non-tax factors. As a result, we
believe that the extent of earnings management that occurred in 1987 in
response to the BIA remains an unresolved issue.
Identifying unexpected accruals: a comparison of current approaches
(Thomas/Zhang forthcoming in Journal of Accounting & Public
Policy).updated 7/25/00.
ABSTRACT
While prior research often uses various accrual
prediction models to detect earnings management, not much is known about
the accuracy, both relative and absolute, associated with these models.
Our paper investigates the accuracy of six different accrual prediction
models, and offers the following findings. Only the Kang-Sivaramakrishnan
(1995) model performs moderately well. The remaining five models provide
little ability to predict total accruals: they are less accurate than a
naove model which predicts that total accruals equal ?5 percent of total
assets for all firms and years. Conventional R2 values from a regression
of actual accruals on predicted accruals are less than zero for a
substantial majority of firms for these five models. These low R2 values
in the prediction period contrast sharply with the much higher R2 values
that are obtained within the estimation period. Similar performance is
observed when predicting current accruals alone. However, the relative
rankings of the different models are altered somewhat: the Jones (1991)
model is then the only model that exhibits some predictive ability.
Positioning liabilities on the balance sheet:
The case of short-term debt that is reclassified as long-term debt
(Gramllich/McAnally/Thomas forthcoming in Journal of Accounting Research).
updated 11/17/98.
ABSTRACT
This is a preliminary investigation of the reasons for and the
consequences of the movement of short-term debt (primarily commercial
paper) between the current and long-term liability sections of the balance
sheet. Under SFAS 6, firms can reclassify short-term debt as long-term
debt if they a) intend to roll over such debt, and b) can show the ability
to refinance that debt (through a loan commitment) if it is not rolled
over. Relative to the management of income statements and other forms of
balance sheet management (e.g. taking liabilities off the balance sheet),
these reclassifications appear innocuous. And yet, there is considerable
cross-sectional variation in the extent to which such reclassifications
occur, and some firms with reclassified short-term debt subsequently
"declassify" that debt by returning it to the current liability section.
We find that the reclassification/declassification decision effectively
smooths the balance sheet by returning working capital (liquidity) and
long-term debt (leverage) ratios to their time-series and cross-sectional
benchmarks. For example, firms reclassifying short-term debt to long-term
debt exhibit contemporaneous decreases in current assets and/or increases
in other current liabilities. We are also able to document that the
amount of debt reclassified is related to a decline in debt and commercial
paper ratings. We explore the potential implications of these findings.
Exchange rate exposure and firm valuation: new evidence for market
efficiency
(Palia/Thomas 1997 working paper). updated 6/1/97.
ABSTRACT
Prior research on the sensitivity of firms to exchange rates has
documented a puzzling result: while firm stock returns are only weakly
related to contemporaneous changes in exchange rates, they are related to
lagged exchange rate changes. This result is consistent with market
inefficiency since the stock market appears to respond to exchange rate
changes with a delay when the predictable effects of dollar movements are
confirmed in financial statements (Bartov and Bodnar, 1995). We re-examine
these results for samples of net exporters and net importers, where net
importers (exporters) are firms with a strong positive (negative)
correlation between abnormal returns and contemporaneous exchange rate
changes over the previous 20 quarters. In contrast with prior results, we
find a strong relation between abnormal returns and contemporaneous
exchange rate changes. We show that our result is due primarily to subsets
of firms and sample periods not examined earlier; specifically, the
relation is strongest for net importers (the international firms examined
by others are more likely to be net exporters) and more recent years.
Although we confirm the lagged relation observed by others, we document
many conflicting aspects of this relationship, and hence are reluctant to
interpret it as market inefficiency.
| 
