Transcript The systematic analysts` forecast errors are predictable

STOCK MARKET LIQUIDITY,
AGGREGATE ANALYST FORECAST
ERRORS, AND THE ECONOMY
Ji-Chai Lin
Ping-Wen (Steven) Sun
Department of Finance
Louisiana State University
Dec 16, 2011
Agenda









Background
Research Question
Conjecture
Hypothesis
Related Literature
Data
Empirical results
Conclusion
Implications
Background
Næs, Skjeltorp, and Ødegaard (2011) find


Stock market liquidity change could predict future
economic state (ex. GDP growth), especially from
small stocks
The flight-to-quality or flight-to-liquidity behavior
from smart investors
Stock Market Liquidity and Business Cycle
Research Question
What is the mechanism in the economic
forecastability of stock market liquidity?
More specifically,
Why are investors willing to buy shares (provide
liquidity) sold by smart investors?
Do analysts play a role?

Conjecture




Smart investors foresee economy change
unexpected by most of the investors
The expectation of most investors is market
expectation
Financial analysts produce information and affect
investors’ expectation
Smart investors foresee economy change
unexpected by “analysts”!
Hypothesis
Information content of analysts’ forecasts
We assume that through aggregation, we can isolate
the systematic analysts’ forecast errors
The systematic analysts’ forecast errors are
predictable

Hypothesis
Flight-to-quality behavior from smart investors
We assume smart investors exploit the systematic
analysts’ forecast errors through trading
Hence, aggregate analysts forecast errors provide
a channel through which stock market liquidity
is a good future economy indicator

Literature Review
Analysts’ information production
Informative?
Givoly and Lakonishok(1979)
Fried and Givoly (1982)
Literature Review
Biased?
 Analysts’ behavior
Dreman and Berry (1995)
Easterwood and Nutt (1999)
Francis and Philbrick (1993)
Clayman and Schwartz(1994)
Lim (2001)
Hong and Kubik (2003)
Literature Review
Security offering & M&A
Ljungqvist et al. (2009)
Kolasinski and Kothari (2008)
 Large and small traders
Malmendier and Shanthikumar (2007)
 Global Settlement
Kadan et al. (2009)

Literature Review
Benefits of Aggregation?
Sadka and Sadka (2009)
Howe, Unlu, and Yan (2009)
Hameed, Morck, Shen, and Yeung (2008)

predictability increases in the process of aggregation.

Aggregation cancels out the idiosyncratic components of analyst
forecasts and isolate their common response to systematic factors

information spillovers as the source of stock return comovement,
where one (neglected) stock is priced using readily available
information about other stocks that share similar fundamentals
Literature Review
Flight-to-quality (flight-to-liquidity) behavior of
investors
Longstaff(2004)
Næs, Skjeltorp, and Ødegaard (2011)
Literature Review
Longstaff(2004)
Flight-to-liquidity premia are related to a variety of
market sentiment measures
 Changes in consumer confidence
 Amounts of funds flowing into equity and money
market mutual funds
 Supply of Treasury securities
Næs, Skjeltorp, and Ødegaard (2011)
Table VII
Major Hypothesis Revisit

Information content of analysts’ forecasts
The systematic analysts’ forecasts are predictable
Flight-to-quality behavior from smart investors
Aggregate analysts forecast errors provide a channel
through which stock market liquidity is a good future
economy indicator

Data
From 1987Q1 to 2010Q4
BEA (Bureau of Economic Analysis)
I/B/E/S (Institutional Brokers’ Estimate System)
CRSP (Center for Research in Security Prices)
 NYSE common stocks (shrcd = 10 or 11) only
 Share price >$1
 Listing for the whole calendar year
Data
Standardized Analyst Forecast Error (SAFE)
SAFE 
QuarterlyactualEPS- ExpectedquarterlyEPSfromanalysts
Standarddeviationof EPSdifferencein theprioreight quarters
December,
Year t-1
March,
Year t
June,
Year t
September,
Year t
January,
Year t
April,
Year t
July,
Year t
October,
Year t
February,
Year t
May,
Year t
August,
Year t
November,
Year t
Q1, year t
Q2, year t
Q3, year t
Q4, year t
Data
Data
GDP growth (logarithm difference)
dGDPRt  ln(GDPRt / GDPRt 1 )
Data
Stock market liquidity
(Amihud (2002) daily price impact)
DT
Ri ,t
t 1
VOL i ,t
ILLIQ i ,T  1 / DT 
Data
Qt-2
Qt-1
Ex:
April, May,
June
Ex:
July,
August,
September
Stock market
liquidity change
Smart investors
Qt
Ex:
October,
November,
December
GDP growth
Aggregate analysts
forecast errors
Data
Control variables
 Market volatility (equally-weighted quarterly
individual stock return standard deviation)
 Market excess return (3-month cumulative monthly
S&P500 index return and risk-free rate difference)
 Term-spread (10-year T-note yield and 3-month Tbill rate difference)
 Default-spread (Moody’s Baa – Moody’s Aaa)
Fama and French (1989)


Default spread is a long-term business condition
variable (high during periods when business is
persistently poor and low during periods when the
economy is persistently strong)
Term spread is a short-term business cycle
variable (tends to be low near business cycle peaks
and high near troughs)
Empirical results
Empirical results
Empirical results
Empirical results
Empirical results
Empirical results
Empirical results
Empirical results
Conclusion


Detrended aggregate analyst forecast errors highly
correlate with concurrent GDP growth
A large part of the forecast errors can be predicted
using lagged macro variables
Conclusion

Once we control for the predicted forecast errors,
the economic forecastability of stock market
liquidity disappears
Conclusion

Stock market liquidity change has virtually no
predictable power on the component of future GDP
growth unrelated to aggregate analyst forecast
errors, suggesting that systematic analyst forecast
errors possess all pertinent information for stock
market liquidity as an economic leading indicator
Implications




Analysts are too optimistic before recessions start and too
pessimistic before recessions end
From the fact that analysts’ forecast errors are predictable,
analysts do not fully take into account of macro variables
(market return, volatility, term spread, and default spread) in
their prediction process
Investors bear an information risk by following analysts’
forecast
Companies may make better investment decisions through a
better prediction of future economy state
Future Research




Industry fixed effects: some industries might be a
better predictor of economic activity than others
Information risk: firms with higher analyst forecast
errors may have higher returns
Smart investors: institutional investors with more
trading, mutual fund managers (stock picking and
market timing)
International comparison: analysts’ quality and
institutional ownership differs in different countries
Thank You!