In the study “Evidence on the Nonlinear Relation between
Insider Trading Decisions and Future Earnings Information,”
University of Chicago Graduate School of Business
professor Darren T. Roulstone and Joseph D. Piotroski of
Stanford University provide new evidence that regulation of
insider trading does impact insider behavior.
In an earlier study, Roulstone and Piotroski established
that corporate insiders (i.e., top officers and directors) are
more likely to sell their company’s stock if future earnings
reports will contain bad news and more likely to buy their
company’s stock if those reports will contain good news: the
relation between earnings news and trading was “linear.”
Given that insiders can be prosecuted for trading on information
not known to other market participants, Roulstone
and Piotroski questioned whether this linear relationship had
a limit. In particular, if future earnings news is highly unusual
and more likely to be visible to the market and market regulators,
would insiders be reluctant to trade on the information?
Roulstone and Piotroski document their results in their
new study.
“The reluctance to trade on extreme news is even more
pronounced when insiders are selling their shares or exercising
stock options,” says Roulstone. “That is consistent with the
fact that when you have really good news to report, no one is
unhappy or cares what you do. But when you have bad news to
report, people focus more on the trading and may file lawsuits.
It’s more difficult to sell before bad news than it is to buy
before good news.”
Good News and Bad News
Prior research on insider trading has focused on whether
insiders are making money trading their own company’s
stock. Roulstone and Piotroski took a closer look at the issue
by examining what types of information insiders have, and
how insiders might use such information. One possible piece
of inside information concerns future earnings.
“Insiders likely know more about how earnings are developing,”
says Roulstone. “Insiders may know many facts about
the firm such as what sales they’re making today that will turn
into income in the future, what products they’re developing,
what projects they’re investing in, and what new markets
they’re trying to open up. All of this information comes
together to give the insider an advantage over outside market
participants in predicting how prices will move in the future.”
Laws against insider trading prohibit insiders from using
material knowledge that investors would consider relevant to
trading stock, and which the insider has a duty to disclose to
market participants, notes Roulstone. The more extreme
earnings news is, the more apparent an insider’s knowledge
of that news will have been, and the more likely the news will
be considered relevant to market participants.
The authors found insiders were particularly reluctant to
keep selling shares and exercising options if future earnings
contain especially negative information.
“If there is eventually a business disaster, it will look bad
for me to sell all my shares right before the stock collapses,”
says Roulstone. “That becomes an Enron-type of situation.
In lawsuits, shareholders often argue that the fact that the
manager was selling shares indicates the manager knew there
was a problem and should have disclosed the news.”
An alternative explanation for the study’s findings is that
extreme news is simply more difficult for insiders to forecast
and trade upon.
“The diminishing effect of extreme news occurs when
something comes in that’s not natural, such as a major change
in the market or the discovery of a new product,” says Roulstone.
“Therefore, you expect to see less of a relationship between
insider behavior today and what the actual news is next year.
The insider simply didn’t see the news coming.”
To investigate this issue, Roulstone and Piotroski measured
the predictability or persistence of changes in company performance.
Persistence was measured by observing whether
a change in annual earnings was followed by a change in
annual earnings in the same direction. A persistent earnings
increase is one that is followed the next year by another
increase. One would expect that the more persistent the increase
or decrease, the easier it would be for insiders to predict.
Thus, if extreme news is less persistent, insiders may not be
trading on the news simply because they cannot predict it�
not because of legal fears. Roulstone and Piotroski separated
earnings changes into persistent and nonpersistent changes
to see if the nonlinear relation (the decline in trading before
extremely good or bad earnings reports) would disappear
among persistent changes.
What the authors found is that even when changes are
persistent, insiders ignore them if they are too large. Thus, the
hypothesis that insiders avoid trading on extreme earnings
changes due to legal risks still holds true: a nonlinear relation
between trading and earnings changes exists even when a
story based on persistence cannot explain it.
The authors found no relation at all between trading and
future earnings news when the future news is nonpersistent
(when increases or decreases in future earnings are reversed
in the subsequent year).
“Our results suggest that insiders are quite successful at
predicting not only changes in financial performance, but
how permanent those changes will be,” says Roulstone. “That
makes sense, because if you are investing based on future
news, you want it to be on news that
will have a strong effect on firm value.”
The information that a company is
going to experience a one-time
change in earnings and then return
to normal the following year is not as
valuable as a permanent change in
earnings-producing capability.
“If you have temporary bad news,
why sell the stock?” asks Roulstone. “But if there is a persistent
decline in firm performance, you may want to start taking
your investments out of your firm. Insiders seem to do a pretty
good job at figuring out which of these shocks are long-lasting.”
Data Use
Roulstone and Piotroski used insider trading data, including
data on option grants and option exercises, from 1996 to 2004
from Thomson Financial, which tracks insiders’ filings with the
SEC. They combined that data with company earnings records
from Compustat and calculated yearly changes in income.
“We took the change in annual earnings divided by current
total assets and related that measure to whether or not there
were buying, selling, or option exercises at the firm,” says
Roulstone. “We measured buying at a firm as cases where
buying of shares exceeded selling of shares among company
insiders. In any given year, people sell stock to cash out their
shares or diversify, even if they don’t know anything about the
future. Similarly, insider selling was measured as cases where
selling exceeded buying among the firm’s insiders.”
Roulstone and Piotroski also accounted for option grants,
since executives tend to sell off some of their prior options or
shares after receiving such grants.
The Value of Earnings
The authors chose to study earnings because they have such
great impact on stock prices, even though they are just one
indicator of a company’s financial state.
“We worry about earnings a lot,” Roulstone said. “When
earnings-per-share numbers come out, we compare them to
analyst forecasts. Stock prices vary wildly based on exceeding
or falling short of forecasts. It’s only one piece of information
that goes into market prices, but it’s probably the piece the
insiders know the best or derive the greatest advantage from�
even more so than analysts. This is a clear advantage the
insider has on one relevant piece of information.”
The authors document that the relation between earnings
and insider trading is not explained by other determinants of
insider trading, such as past stock returns or the firm’s market-
to-book ratio. However, the study’s findings regarding
earnings should not be overstated, says Roulstone.
“Over the course of a year, earnings are a very broad measure,”
says Roulstone. “A lot of other financial information about a
company comes out, and often much quicker than earnings.
Earnings are a lagging indicator of how a firm is doing, while
stock prices usually anticipate the public disclosure of an
earnings change. But regardless of past, present, or future
returns on their company’s stock, insiders buy or sell their
shares in anticipation of future earnings reports. They have
an advantage regarding earnings news and they exploit it.”
"Evidence on the Nonlinear Relation between Insider Trading Decisions and Future Earnings Information." Joseph D. Piotroski and Darren T. Roulstone. Forthcoming in the Journal of Law, Economics, and Policy.