Ivo Welch, Brown Economics
January 2009

The Results of the Equity Premium January 2009 Survey

This page describes the answers. The resulting data set itself, sans identifying information, can be downloaded eqsurvey-post.csv. If you really need to cite the answers from this survey, please use 

   Welch, Ivo. "Views of Financial Economists On The Equity Premium And Other Issues," The Journal of Business 73-4, October 2000, 501-537, with 2009 update

In brief, the results of the January 2009 survey are as follows:

Here are detailed univariate statistics for each of the questions:

Short Academic Equity Premium Survey for January 2009

I know I promised not to rerun this survey more than once every few years, but the events of 2008 have made it too tempting for me not to run this again.

Your answers to this short survey will be used to update my Journal of Business equity premium survey from 1998 (and its follow-ups from 2001 and early 2008). Your answers will be held strictly confidential. I will not publish the results in an academic journal, but I will post them on SSRN and on my own website in working paper format. [I am doing this as a public service, not to garner a line item on my CV.]

If you have difficulties filling out this survey, please send an email to Ivo Welch.

(Please note that I will pay most attention to US researchers, because these questions are all about the US markets.)

Personal Information

My email address is:
I am a finance or economics professor: Yes = 143 (I only use yes below)
No= 6
NA= 3
Relative to other financial economists, I would guess that I have thought about the equity premium No Answer = 2.
a lot more carefully = 14.
more carefully = 40.
about the same = 49.
less carefully = 27.
a lot less carefully = 11.

Relative to my views 12 months ago (i.e., after the sharp decline in 2008), my view about the future stock market's long-term performance today relative to one year ago is:

 no answer = 1
a lot more bullish = 10
more bullish = 42
about the same = 61
more bearish = 24
a lot more bearish = 6

Background Information: For the prevailing yield curve, click here. For the prevailing Dow-Jones, click here.

My Knowledge of Today's Status

As I am filling out this survey, I know that the Dow Jones 30 stands at roughly (or "exactly," or leave blank if you don't know): Median = 8,200

Relative to other finance professors answering this survey, I believe that I am

 no answer = 16
a lot more bullish = 5
more bullish = 35
about the same = 68
more bearish = 17
a lot more bearish = 2

Parametric Stock Market and Equity Premium Estimates

I expect the average nominal geometric stock return (not equity premium!) over the next 30 years to be Median = 8% (Mean=8.3%). Sd= 2.65%. Q1=7%. Q3=10%. .
I expect the average equity premium over the next 1 year to be
(define avg equity premium as the expected return on the value-weighted US market net of short-term (3 month) T-bills)		 Median = 6% (Mean=6.2%). Sd= 6.4%. Q1=4%. Q3=9% .
I expect the average arithmetic equity premium over the next 30 years to be
(relative to rolling future contemporaneous short-term (3 month) T-Bills*) 		Median = 6% (Mean=6%). Sd= 2.2%. Q1=5%. Q3=7% .
I expect the average geometric* equity premium over the next 30 years to be
(relative to rolling future contemporaneous short-term (3 month) T-Bills)		 Median = 5% (Mean=5.1%). Sd= 1.9%. Q1=4%. Q3=6% .
G30-A.1: Same question: In your classes, what is the main number you are recommending for long-term CAPM purposes? Median = 6% (Mean=6.2%). Sd= 1.7%. Q1=5%. Q3=7% .
G30-A.2: Same question: In your classes, if you give a reasonable range for CAPM use, what is it? Median = 4% (Mean=4.75%). Sd= 2.7%. Q1=3%. Q3=6% to
Median = 7% (Mean=8.2%). Sd= 5.6%. Q1=6%. Q3=8.25% .

Non-Parametric and Probability Equity Premium Estimates

Please give me an over/under bet for the Dow Jones for December 31, 2009:
your level estimate should result in a risk-neutral, fair bet for either side (i.e., not adjusted for hedging/risk premia)		 Median = 8,825 (Mean=8,644). Sd= 1,529. Q1= 8,500 Q3= 9,138
What is the probability that the stock market will go down over the next 12 months?
think of the market here as the rate of return on the Dow-Jones (level plus dividends) total rate of return, not the equity premium. 		> Median=35%. Mean=35.8%. SD=18%. Q1=25%. Q3=49% probability
What is the probability that the stock market will decline (lose money = nominal) over the next 12 months by 20% or more? > Median=10%. Mean=15%. SD=15%. Q1=5%. Q3=20% probability
What is the probability that the stock market will decline (lose money = nominal) over the next 10 years? > Median=5%. Mean=7.7%. SD=11%. Q1=1%. Q3=10% probability

Current Interest Questions

Leave blank if uninterested.

How much in fiscal stimulus should the Obama government expend? > Median= $700 billion. Mean=$684. SD=$625. Q1=$100. Q3=$1,000 billion dollars
Was it necessary or useful for the government to bail out the financial sector (rather than letting the financial institutions fail)? no answer = 20
yes = 93
no = 20
Has the Bush government spent the first half TARP money wisely? no answer = 26
yes, they did a good job = 9
yes and no, they did so so = 66
no, they did a bad job = 42
Should the government force banks that receive TARP money to stop paying dividends to equity? no answer = 25
yes, the government should prohibit dividend payments = 86
no, the government should allow dividend payments = 32
Should the government force the subordinated debt of banks that receive TARP money to be converted into equity? no answer = 29
yes, the government should force a debt-to-equity conversion = 64
no, the government should not force a debt-to-equity conversion = 50
Does corporate governance need to be reformed (my GAAP-like suggestion)? no answer = 32
yes, corporate governance is broken and needs some reform = 60
yes, corporate governance is broken and could use a GAAP like agency = 17
no, corporate governance should not be touched by any further laws or statutes = 28
no, corporate governance would be better if all laws or regulatory statutes were removed = 6

This Survey

Were the questions in this survey clear? no answer = 10
clear = 106
muddy = 25
unclear = 2

Do you want me to email you with the results when I have them? no answer = 9
yes = 127
no = 7

* Note that different people use different equity premium definitions. I am asking it relative to short-term (3 month) Treasuries, not relative to long-term Treasuries. If you are used to quoting an equity premium relative to long-term Treasuries, you should add a term-spread estimate. Also, the geometric equity premium is "casual" usage. Think of it as compounded equity return minus compounded risk-free.