Empirical Asset Pricing 35905

Prof. John H.Cochrane

john.cochrane@chicagobooth.edu HPC459 702 3059

Latest update: Feb 28 2011 .

NOTE: some of the links on this page have broken with the passage of time. I'll fix them if and when I teach the class. In the meantime, if you find a broken link just google the paper instead.

Class Description
Readings
Documents - problem sets, lecture notes, etc.
More

Announcements

Friday problem sets
1. Send it by email rmano@uchicago.edu 2. leave it in Rui's folder in Rosenwald 2nd floor, by 3.30pm

We WILL have a class Mon Jan 17. Really, do you want to try to reschedule? You will need a booth id to get in the door. Rui will be by the main door 5807 S. Woodlawn 8:20-8:30 to let in anyone who doesn't have a card.

Class Meeting Times: M 8:30-12:00 C09

Review Wed 8:30-10 3A

No friday class. We'll do a three hour class on mondays. There was no way to get around conflicts

Exam: Finals week, C09 M 8-11 Per booth exam schedule

Class description

Philosophy

First I survey facts, then theories constructed to understand the facts.

Mechanics

The class will center on reading and discussion of articles and problem sets. You must read and think about the readings before class, and be ready to discuss the readings in class. I will call on you and occasionally ask students to lead discussions on papers or parts of papers. You should be ready to do this.

Grades will reflect class participation, homework, and a final exam. The final exam takes place as per the Booth final exam schedule.

Bring a name card to every class if you want class participation grades or me ever to learn your name.

Write your own problem set. You may talk to anyone you want while doing problem sets.

You will need a copy of my Asset Pricing, preferably the revised edition with no (known) typos. The articles will be available as pdfs from the class website. You need to be able to do empirical work. I recommend matlab and can help with it, but use what you want. You need to be able to get crsp data from wrds. Figure out how to do it.

Prerequisites: I don't enforce strict prerequisites. I presume everyone has taken either my or Ralph Koijen's 35904 or George Constantinides' 35912, and a large part of the first-year PhD micro and macro sequences. If you haven't taken finance before, at least really read Ch1 of Asset Pricing carefully before class starts.

Rules: Note some problems are similar to previous year's problems. Using previous year programs or solutions is a major honor code violation. There are enough changed numbers, changed questions, typos and mistakes in previous solutions, that it will take more time to use those than to start fresh.

Class Documents - problem sets, lecture notes, etc.

Problem set 1 - due Mon week 2
Problem set 1 answers Yes, I meant "consistent" not "unbiased".
Program
Olsgmm function called by program
Data files: wvr sbbi cay vwr and sbbi updated 12/28/2011. Note my program is set up with data files in a different directory. Modify the load statements according to where you put data.

Problem set 2 - due Mon week 3
Problem set 2 answers Program

Problem set 3 questions. Due Friday week 4
Problem set 3 answers Program Also uses vwr from above

Problem set 4 1/29 questions Due next Friday
Problem set 4 answers Program

Overheads The pictures I showed in class. Updated 1/23/2011

Lecture notes. On stock predictability. Beginning through the dog that didn't bark. Updated 1/23/2011. Cleaned a bit, and includes the way I presented things in lecture 2011.

Disclaimer: These are my notes. They are not a substitute for doing the readings, and they certainly are not a substitute for showing up in class! They will help you if you miss a point in class. They are not written well. I also massively over-prepare, so I will not cover in class every point made in the notes.

Bonds and FX overheads 2/7/2011 update, reflects class 2/7,
Bonds notes 2/7/2011 update,

FX notes 2/14/2011

Problem set 5 questions At last! Worth the wait.
Data: famabliss.txt Problem set 5 Answers Program

Fama-French Overheads 2/21

Empirical methods notes. Time series, cross section, and eigenvalue decompositions 2/21

Problem set 6 2/21
Factors data; portfolio returns data 2/21 Problem set 6 answers 2/28
Program
tsregress2 function

Problem set 7 questions.2/28
Size data Beme data (Also use Fama French 25 portfolios and factors above)
Problem set 7 answers; Matlab program; Other beme data used in my program 3/7

Notes on Frictions and Financial Crisis. From Lamont&Thaler to Duffie 3/7

---- End of active Class Documents links --------totaldiff function

Fama-French notes 2/25 update includes empirical methods notes,and a derivation of eigenvalue factor models by popular request

Production notes 3/3 update3/8

Frictions and financial crisis notes 3/10
Slides on financial crisis (lots and lots of pretty pictures) 3/10

Final exam review problems 3/10 Revised 3/13 to fix the mistake in the answers to stock predictability question 1

Final exam answers 3/18

Notes on equity premium and lightning m* review
Notes on International risk sharing and Habits
Notes update habit notes, and notes on utility functions
Notes on investment and production

Readings

I expect you to read the "main" readings closely. There are some "reference" readings which contain material I'll talk about in class and additional papers you should know about, but I don't expect you to read them.

Week 1.

  • Discount Rates; talk and slides. The "talk" is the speech I'm giving at the AFA meetings, "Discount rates" is a draft of the underlying paper. It's also an outline of pretty much everything we're goinig to do this quarter.

Week 2-3: Return predictability, long horizons, volatility, bubbles, present value relations, price and return decompositions, VAR representation, univariate mean-reversion.

Reference

I have only assigned the "summary" version of predictability, but of course I hope someday you'll sit down and look at some of the originals, especially mine.

John Campbell Classics: The linearized present value relation, and some of his contributions to the stock predictability literature. (There's lots more on Campbell's website.)

  • Campbell, John Y. and Robert J. Shiller, "Cointegration and Tests of Present Value Models" Journal of Political Economy 95, 1062–1088, October 1987.

  • Campbell, John Y., and Robert J. Shiller, "The Dividend-Price Ratio and Expectations of Future Dividends and Discount Factors", Review of Financial Studies 1:195–228, Fall 1988.

  • Campbell, John Y., and John Ammer, "What Moves the Stock and Bond Markets? A Variance Decomposition for Long-Term Asset Returns", Journal of Finance 48:3–37, March 1993.

Week 4a: Econometric issues in return predictability.

Week 4b Bond return predictability

  • Cochrane, John H., Asset Pricing Ch. 20.1 "Bonds" and "Foreign Exchange" p. 389-435.

  • Cochrane, John H. and Monika Piazzesi, Bond Risk Premia March 2005, American Economic Review 95:1, 138-160. Appendix ; Overheads

  • Cochrane, John H. and Monika Piazzesi 2008, Decomposing the Yield Curve, Manuscript. This paper merges what we learned about predictability with an affine model.

Reference

  • Fama, Eugene F., 1986, "Term Premiums and Default Premiums in Money Markets." Journal of Financial Economics 17, 175-96. There are amazingly few regressions analyzing credit spreads. How much time-series and cross-sectional variation in credit spreads reflects default, and how much measures risk premium? It's a good 2nd year paper topic!

  • Francis A. Longstaff, Kay Giesecke, Stephen Schaefer and Ilya Strebulaev. (March 2010). "Corporate Bond Default Risk: A 150 - Year Perspective.'' Some basic facts, spreads correspond to returns not default. But there are lots of defaults!

Week 5a FX predictability

  • Cochrane, John H., Asset Pricing Ch. 20.1 "Bonds" and "Foreign Exchange" p. 389-435.

  • Lustig, Hanno, Nikolai L. Roussanov, and Adrien Verdelhan, 2009, "Common Risk Factors in Currency Markets," Manuscript UCLA. Up through p. 15 only. This is quick but cool. Rather than run regressions, sort in to portfolios and look at means; then eigenvalue decompose the portfolio covariance matrix. It makes the connection between regression and Fama-French procedures.

  • Jurek, Jakub, 2008, "Crash-Neutral Currency Carry Trades," Manuscript Princeton University. This paper also goes on and on, so don't try to read it all. I will show in class Table I, II; VI-IX, ; Figure 1, 3, 8. These update the UIP regressions, show you what happens if you add put option protection, and show what happened in the big crash.

Reference

NEW SCHEDULE

Week 8 Cross-sectional facts Size, B/M, momentum, accounting sorts, in expected returns and covariances

  • Cochrane, Asset Pricing , Ch. 20.2 p. 435-454.

  • Fama Eugene F. and Kenneth R. French 1996 "Multifactor Explanations of Asset Pricing Anomalies," Journal of Finance 51, 55-84. Really understand Table I, sales rank in Table II, II, Table VI and VII. Also read p. 75 on, "interpreting the results."

  • Fama, Eugene F., and Kenneth R. French 2006, "Dissecting Anomalies" Journal of Finance 63 (4) 1653-1678. Understand the tables. These two are probably the most important papers in all asset pricing!

  • Asness, Cliff, Toby Moskowitz and Lasse Pedersen, June 2009, "Value and Momentum Everywhere" Manuscript, University of Chicago. The latest on value and momentum, and how they are correlated across countries and asset classes.

Reference

  • Lettau, Martin and Sydney Ludvigson, 2001, Resurrecting the (C)CAPM: A Cross-Sectional Test When Risk Premia Are Time-Varying The Journal of Political Economy, Vol. 109, No. 6 (Dec., 2001), pp. 1238-1287. The first figure says it all, no need to go much further. There is an industry that prices the FF 25 with ICAPM or macro models. Often the papers are a bit fishy – huge factor risk premia, small spread in betas, 10 factors to explain 3 premiums. I'll just show Figure 1 here and make a few comments. But I wanted some placeholder for this massive literature. (We since figured out that pricing the 25 isn't very interesting.)

  • McCloskey, Deirdre (then Donald), 1983, ,The Rhetoric of Economics 21, 481-517. This is the article I mentioned that helps to understand why Fama and French are so successful despte needing exponential notation to describe probability values. Easy but very provocative reading.

Week 9/10a Short sales and liquidity; securities with value in trading; order flow and prices

We just had a huge financial crisis. We can only scratch the surface, but this will frame so much research for the next few years I can't leave it out. Liquidity, bank runs, "flight to quality," short constraints, funding constraints, mortgages, regulatory arbitrage, the idea that stock prices could be driven down when hedge funds were forced to liquidate, lots of little "arbitrages," including different prices for bonds and CDS, violations of covered interest parity, a huge on the run/off the run spread, the idiocy of writing put options and calling it "arbitrage," and so on. This should be a a whole course, and I'm sure some day it will be.

Many (most) of the facts and ideas springing from the financial crisis are not new, but just bigger versions of old puzzles. For this reason, we'll read some pre-crisis papers that bear on the same issues.

  • Lamont Owen, and Richard Thaler 2003, Can the Market Add and Subtract?: Mispricing in Tech-Stock Carve-Outs Journal of Political Economy 111: 227-268. A really nice mis-pricing. Two ways of getting 3-com stock trade at different prices.

  • Cochrane, John H., Stock as Money: Convenience Yield and the Tech-Stock Bubble in William C. Hunter, George G. Kaufman and Michael Pomerleano, Eds., Asset Price Bubbles Cambridge: MIT Press 2003. An answer to Lamont and Thaler. Money and bonds are also two ways to get the same thing at different prices, and they look a lot like the two ways of getting 3 com stock.

  • Brandt Michael and Kenneth A. Kavajecz, 2004, Price Discovery in the U.S. Treasury Market: The impact of Orderflow and Liquidity on the Yield Curve Journal of Finance 59, (Dec) 2623-2654. (For class, you can stop reading at section III p. 2644). This is a great paper. The price of A changes when there is order flow of B, suggesting that information leads to orderflow, not orderflow mechanically moves prices.

References.

I'll present results from these in class.

  • Francis A. Longstaff 2004, "The Flight to Liquidity Premium in U.S. Treasury Bond Prices Journal of Business 77, 511-526, 2004, Contrasts treasury bonds with identical Refcorp bonds to isolate the liquidity premium. Fall 2008 saw the same sort of thing 10 times larger.

  • Mattias Flecenstein, Francis Longstaff and Hanno Lustig, 2010, "Why does the Treasury Issue Tips?" A huge arbitrage between treasury and tips, illustrating perhaps some liquidity issues. Read the description of the trade, see Figure 1

  • Arvind Krishnamurthy and Annette Vissing-Jorgenson, 2010, "The Aggregate Demand for Treasury Debt" Just look at Figure 1. There seems to be a "money" like demand for all treasury debt.

  • Cochrane, John, 2004, Liquidity, Trading and Asset Prices, NBER Reporter This is an overview paper. It gives my organizing thoughts on both theory and empirical work and covers some papers we won't read directly.

Week 10b Financial Crisis

  • Brunermeier, Markus, 2009, Deciphering the Liquidity and Credit Crunch 2007-2009Journal of Economic Perspectives 23 77-100. (Optional longer version NBER working paper with more graphs). I will only cover the facts and chronology.

  • Gorton, Gary B., and Andrew Metrick, 2009, Haircuts, NBER Working Paper 15273. The heart of the "crisis" was a run on short term debt, and the fact that suddenly all sorts of debt was not accepted as collateral. Read about it here. This is also excellent on how short-term debt works, and offers a well-explained alternative and diametrically opposed view to the Diamond-Rajan view that short-term debt is necessary to discipline managers. Here, short-term debt is useful precisely because you do not have to do any monitoring.

  • Duffie, Darrell, 2009, The Failure Mechanics of Dealer Banks  Journal of Economic Perspectives 24, 51–72. This is fantastic, both to understand nuts and bolts, why there were runs, and how these markets work. If you want to know what the "fail" in "too big to fail" means, read here. It's since grown to a book from Princeton University Press.

Reference Some of my writings on financial crisis

  • "A Skeptical Appraisal of Frictions in the Financial Crisis" September 2010. Notes and Pictures. A skeptical view of the emerging consensus that the financial crisis is all about "bubbles" "liquidity spirals" "fire sales" "capital constraints" at commercial banks and so on. I do think there was a run on repo and short term financing. Good old fashioned macro asset pricing works a lot better than you might have thought.

  • Asset pricing after the crash  March 20 2009 This is a piece based on a panel discussion titled Rethinking asset pricing at the Spring 2009 NBER Asset Pricing meeting. It includes sceptical views on just how important credit constraints and liquidity really are. Liquidity is the frosting on the cake of finance. There is a lot of frosting these days, but still some cake.

  • Lessons from the financial crisis Jan 2010 Regulation 32(4), 34-37. The financial crisis is mainly about too big to fail expectations. The only way out is to limit the government’s authority to bail out.   

Week 10c Closing thoughts. Fun, and speculating about future research

Additional Reading Suggestions (i.e. things I couldn’t stuff in to 10 weeks)

Equity premium

Habits

References

Preferences and macro-asset pricing

A quick overview of the other preference-based approaches to macro/asset pricing, recursive utility, incomplete markets, multiple goods, labor income. These topics are far more important than the short treatment I'm going to give them, but I'm leaving the big work to Hansen and Heaton.

Dynamic Portfolio Theory.

  • Cochrane, John H., Portfolio theory p. 1-15; (feel free to read 16-39, but the next reading does it better.) p. 40-79;

  • Cochrane, John H., "A Mean-Variance Benchmark for Intertemporal Portfolio Theory" Feb 2008.

  • Cochrane, John H., Manuscript for Wall Street Journal Op-Ed. Dynamic portfolio theory matters enormously, in real-world practice. "Hedge demands" are not an abstract curiosity. Certain big endowments that sold at the bottom in Dec 2008 should have known better. John Campbell pointed out a grey area in my analysis. It’s not obvious that less risk-averse people will buy assets from more risk averse people after a market crash.

References

Liquidity, shorting, downward sloping demand

These are more general for if you get interested in the subject

  • Yakov Amihud, Haim Mendelson, and Lasse Heje Pedersen (2005), "Liquidity and Asset Prices," Foundations and Trends in Finance, 1, 269-364. A good survey of the literature.

  • Lamont, Owen, (2004) "Go Down Fighting: Short Sellers vs. Firms" Manuscript, Yale University . Great stories and some remarkable alphas showing firms with short constraints are "overpriced."

  • Lamont, Owen, Short Sale Constraints and Overpricing Manuscript Yale University. An easy-to-read summary

  • Darrell Duffie, Nicolae Garleanu, and Lasse Heje Pedersen (2002) "Securities Lending, Shorting, and Pricing,", Journal of Financial Economics, 66, 307-339.A model of the "liquidity effect" explanation of short sales mispricing like 3Com/Palm

  • Martin D.D. Evans and Richard K. Lyons 2002, "Order Flow and Exchange Rate Dynamics" Journal of Political Economy, 110: 170-180. One of the first to correlate order flow and price changes; the puzzle that Brandt ties up nicely. For us, the source for the graph I'll show in class.

  • Shleifer, Andrei, 1986, "Do Demand Curves for Stocks Slope Down?" The Journal of Finance, 41, 579-590. S&P500 inclusion gives a tiny price blip. Historically important; one of the first to show that they do, a bit.

  • Nicholas Barberis Andrei Shleifer and Jeffrey Wurgler, 2005 "Comovement Journal of Financial Economics, 75, 283-317 This follows up on Shleifer's original, showing stocks start to move together more when they get included in the SP500. Before you get too excited, notice how small these effects are.

  • Mitchell, Mark, Todd Pulvino and Erik Stafford, 2004, "Price Pressure Around Mergers Journal of Finance, 59, 31-63. When A tries to buy B, a lot of traders try to buy B and short A. They claim that this lowers the price of A.

  • Mitchell, Mark , Lasse Heje Pedersen, and Todd Pulvino. 2007. "Slow-Moving Capital." American Economic Review, 97(2): 215–220. DOI:10.1257/aer.97.2.215.Slides

  • Pastor, Lubos, and Robert F. Stambaugh 2003, "Liquidity Risk and Expected Stock Returns" Journal of Political Economy 111, 642-685. Does cov(R, liquidity) drive expected returns?

  • Acharya, Viral V. and Lasse H. Pedersen, 2005 "Asset pricing with liquidity risk" Journal of Financial Economics, 77, 375-410 This one estimates the four kinds of correlation of price with liquidity. The "theory" assumes people live two days, a convenient but obviously, er, simplified motive for trade.

"Production based"' and investment models.

Reference: I will discuss in class some results from the following

I'm leaving out here the very important topic of general equilibrium models that address the asset pricing facts (e.g. Gomes, Kogan, Zhang; Gala; Gourio). Again, we have to stop having fun somewhere. Getting value and growth effects out of models like these has been a big challenge. You need an interesting cross section of firms, but you don't want to carry around huge state variables .

Financial crisis

There is a huge amount being written on the financial crisis of course. Anil Kashyap will teach an MBA course on it this spring, which you might take if you want to do research on this topic. These struck me as particularly good

General:

  • Krishnamurthy, Arvind, 2008, "Fundamental Value and Limits to Arbitrage," Manuscript, Kellogg Graduate School of Management. A very nice short documentation of two "arbitrages" from the crash. I'll show some more pictures, including the on the run/off the run, covered interest parity, SPAC and CDS-Bond spreads.

  • Cochrane, John H, Lessons from the financial crisis Jan 2010 Regulation 32(4), 34-37. In my view, solving TBTF expectations is the crucial policy challenge, and "regulation" with a broader TBTF guarantee isn't going to do it.

  • Cochrane, John H., Asset pricing after the crash Includes skeptical views on just how important credit constraints and liquidity really are. Liquidity is the frosting on the cake of finance. There is a lot of frosting these days, but still some cake. (A response to Brunnermeier)

  • Cochrane, John H., A Skeptical Appraisal of Frictions in the Financial Crisis September 2010. Notes and Pictures. This is a 2 hour lecture for Ph.D. students at the Deutsche Bank Symposium hosted by the Booth School September 2010. It offers a skeptical view of the emerging consensus that the financial crisis is all about "bubbles" "liquidity spirals" "fire sales" "capital constraints" at commercial banks and so on. I do think there was a run on repo and short term financing. Good old fashioned macro asset pricing works a lot better than you might have thought.

  • Swagel, Phillip, 2009,"The finanical crisis: An inside view" Forthcoming Brookings Papers on Economic Activity. This is a nice view of the political and legal constraints as well as the government's perspective. Not really asset pricing, so not on the main list, but if you want to understand why the government did apparently crazy things, go here.

Arbitrages:

Regulation:

Structured products and repos

  • Gorton, Gary, 2008, "The Subprime Panic " Manuscript, Yale University, forthcoming European Financial Management. This has a good summary of how subprime mortgages and complex structures actually worked. I don't totally buy the "obscure assets" and information story, that the ABX index forced marks to market, but it's well worth reading.

  • Coval, Joshua, Jakub Jurek and Erik Stafford, The economics of structured finance, Journal of Economic Perspectives, 23:1, 3-26. A nice piece showing how tranches of securitized debt end up creating index options – same probability of default, more systemic and hence higher priced risk.

  • Gorton, Gary B., and Andrew Metrick, 2009, "Securitized Banking and the Run on Repo" Yale ICF Working Paper 09-14 Only read to p. 17. The "run on repo" description is good, the regressions are weaker.

  • Covitz, Daniel M., Nellie Liang and Gustavo Suarez, 2009, "The Evolution of a Financial Crisis: Panic in the Asset-Backed Commercial Paper Market" Manuscript, Federal Reserve Board. Empirical work in the asset-backed commercial paper market showing how ABCB is prone to "runs" like old-fashioned bank accounts.

  • Krishnamurthy, Arvind, 2009, "How Debt Markets Have Malfunctioned in the Crisis" Manuscript, Kellogg Graduate School of Management. Forthcoming Winter 2010 Journal of Economic Perspectives. A simple overview of repo and other debt markets, and some little arbitrages. I don’t buy the "spirals" but the description is good

Lite:

  • Cochrane, John H., The Monster Returns Will the silly idea of buying up "toxic assets" not go away?

  • Did "Efficient Markets Cause the Crash," and does the crash prove behavioral was right all along? Fama and French refute this silliness on their blog, and Ray Ball wrote a beautiful essay, and I have a few choice words in a longer response to Krugman's New York Times article on the crash.

  • Berkshire Hathaway Letter From mid 2008. This is beautifully written and I agree with almost all of it, and illustrates Buffet’s thinking in the depth of the crisis. Buffet’s views on writing insurance are a real high point. Part of my "put option" writing sermon was that writing put options is a good business, if you are transparent about the risks. Buffet is. However, he makes a great point that municipalities are much more likely to default if they know an insurer will lose money rather than if their own taxpayers and voters will lose money, so that historical default probabilities are a poor guide. The only place he gets it a little bit wrong is in the last diatribe against the Black-Scholes formula at the end. He forgot about "state prices." A parachute, delivered to row 3b in the event that the back of the plane blew up, is worth a lot more than the $1000 it costs to order on line. Similarly, the value of any promise to pay something in the state of the world that the S&P is lower 99 years from now than today (essentially, the economy is back to the stone age) is worth a lot more than probability x dollar suggests.

----- THE END-----