Should I Read Academic Research On Finance?

ERcaco's picture
Rank: Senior Monkey | 82

This week, a prominent American Economist famed for his work in Behavioral Finance, Richard Thaler, took home the Nobel prize in Economic Sciences. As I found myself reading these headlines in the news I found some pride in being able to say I was already familiar with his work and the work of his contemporaries.

In fact, I recently made an effort to start reading more scholarly articles on finance.

Some of the papers I read include:

  • A paper that claims most research findings in financial economics are false due to data mining.
  • A counter paper which argues that fundamental signals predict cross-sectional stock returns even after accounting for data mining AND that abnormal returns cannot be attributed to random chance and are better explained by mispricing.
  • Lastly, a paper that states GAAP Earnings have significantly lost their value in predicting abnormal returns since the 1980s (which I posted a discussion on here).

After reading these papers and seeing some of the comments made on my discussion post, I wonder if anyone working in the industry even pays attention to what is written in academic research papers. Or, perhaps I'm wasting my time and should pursue other areas of knowledge more valued by practitioners.

So, my questions to you...

Do you ever read academic research on finance and apply it to your work? If so, is there anything you would recommend?

Comments (32)

Oct 12, 2017

Based on my experience I can say people on the Quants team in my bank do read a lot of academic papers.

"The markets are always changing , and they are always the same."

Oct 12, 2017

Hi JoyfulMonkey!

Thanks for taking the tome to comment. In regards to the question, do you mind if I ask what function you work in?

Jan 9, 2018

Currently I am an intern at control team.

One of my tasks is to summarize the documents on new valuation methods developed by the Quants team in London, for the FO, who base their recommendations based on academic papers.

"The markets are always changing , and they are always the same."

Best Response
Oct 12, 2017

I read a lot of academic research/academic broker papers at work for my job (quantitative trader) - they are important things to read to come up with a successful trading strategies. I'm not sure who gets value from getting a research paper from a brokerage telling them that bonds up, credit spreads are down etc - useless.

To your point regarding "most research findings in financial economics are false due to data mining" is a very valid point. When you read a research paper you have to think of it as a traditional scientific experiment which goes through the scientific process. When you create your hypothesis - typically the hypothesis has to be logical to you to some degree. For example consider these two examples, buy stocks every Thursday and sell them on Monday versus buying stocks a day before end of the month and selling stocks on the last day of month to trade ahead of pension contributions/401k contributions/institutional re balances etc. Which one of these makes more sense? When you think about conducting your experiment you can question why Thursday and Monday? why not Wednesday and Tuesday? versus end of the month where you have a specific reason which is influenced by external economic forces. After you have your hypothesis you have a concrete example to test. Going back to the first example many researchers will say well my experiment failed but I need to get something out - and my infrastructure is already in place so I'm going test all xyz combinations until one makes money and that's the one I publish. And that is going to be the first one I found. Additionally some research papers will find that trading in front of contributions generates no alpha (valid conclusion and this is an important place to reach to know what doesn't work) - but in the end say but if you trade on the 15th of the month - this will work and run a new study there (this on the other hand is not part of the hypothesis and should be discarded based on no foundational thought).

In short academic papers are important but it requires some skill and philosophical debate to separate out ones are useful and ones that due to date mining. If everyone could generate alpha in their papers that would make our jobs too easy.

    • 9
Oct 12, 2017

Thanks @Sasquatchosaros

I think i see what you mean. Essentially, if I look long enough for new factors, I'm bound to find one. But, it is very likely that any correlation between alpha and this new factor is spurious if my initial assumptions are not well founded?

Oct 12, 2017
Jcvcaco:

Thanks @Sasquatchosaros

I think i see what you mean. Essentially, if I look long enough for new factors, I'm bound to find one. But, it is very likely that any correlation between alpha and this new factor is spurious if my initial assumptions are not well founded?

That is correct - start with a hypothesis and work down to a result. Rather than start with a result and try to explain it with a hypothesis.

    • 2
Jan 9, 2018

AQR publishes good stuff. So do a bunch of the other asset management firms. Damodaran's blog (which appears on this website) is also very good. He is a very clear thinker; cuts through a lot of the nonsense that surrounds this field.

Jan 9, 2018

Many thanks, but isn't AQR's stuff only available to the fund's investors?

Jan 9, 2018

Props for CQS mention.

Jan 9, 2018

If anyone has sources to add, just do a quick shout and I will add stuff to the OP.

Jan 9, 2018

I am a big fan of the stuff both Jeremy Grantham and James Montier at GMO write. Even though they don't do it frequently, they do produce really good stuff.

Jan 9, 2018

D-OH! Sorry, just noticed that these were mentioned already in the original post...

Jan 9, 2018

Sequx investor day transcripts and annual reports are good reads. I also check marketfolly regularly anc the securityanalysis subreddit.

Jan 9, 2018

Old Man Marley. (placeholder - curious what else turns up).

Jan 9, 2018

Home Alone fan or am I missing something?

Jan 9, 2018

Where do you find the GMO letters on their website? I can find with Google, but would prefer finding a more consistent source to keep tabs on.

Also, does AQR have the most impressively educated principals or what? Holy shit..

Jan 9, 2018

the CFA institute sends me magazines with papers in it. i havent read any of them

Jan 9, 2018
bluegold:

the CFA institute sends me magazines with papers in it. i havent read any of them

lol thanks for making my lunch break. There's usually a couple each year that are worth skimming but otherwise you're right!

Jan 9, 2018

May not fit into this list but I've been reading "Margin of Safety" by Seth Klarman recently. Its hard to find but you can get a PDF off the net if you try hard. I would be willing to email it around if someone really wants it.

Oct 13, 2017

IF you can understand them first of all and then if you have the inclination. I see no harm?

Oct 13, 2017

Yes. You should. Odds are that most of your peers don't read these papers and, in turn, you can gain a one-up on some of them. I would also highly recommend reading tons of academic literature outside of finance. I have found that some of my best investing and personal insights have come from philosophy, sociology, psychology, evolution, etc. I think a great example applying concepts outside of finance to investing comes from Peter Thiel's investment in Facebook (certainly, one of the best investments of all time). "Thiel's inspiration for investing in Facebook was Girard's mimetic theory" Link

    • 1
Oct 15, 2017

I think I would read academic papers if I became more interested in algorithmic trading. It seems to me that having a really firm grasp of statistics is paramount and academics are supposed to be pretty good with stats.

I also like to keep an open mind about things and from that standpoint I would not recommend ignoring academic research. But my (and your) time is limited.

Is it a better use of your time to read a bunch of earnings calls, investor days, 3rd-party research, SEC filings and industry information to get your informational edge? For me I think it is. For most people in fact, I would think it is.

What is someone hoping to gain from reading academic research? Be careful to not fall into the trap of reading something just to say that you read it, to sound smart, to feel like you are on the cutting edge. If you want to outperform, all that matters is that you actually have that edge.

Now, there are other ways to make a living (beyond outperforming) and some of those ways involve reading academic research and sounding smart. That could be your thing.

    • 1
Oct 15, 2017

I read academic papers in my position (research). I also go to lectures, but most of these are targeted at a professional audience, rather than academics themselves.

Oct 22, 2017

Thank you @mbamfinquestions

Would you mind sharing which lectures you like to attend? The lectures you mention sound particularly interesting because they are targeted at a professional audience. I imagine they (the lecturers) emphasize how to apply the concepts.

Oct 27, 2017

I go to ones that at are typically invitation-only. Some fund managers organize lectures and presentations for affiliate parties, so those are most of what I attend. The CFA society also coordinates some stuff. The danger is that these lectures can be pitched at a low base. All of these lectures are different to what undergrad lectures were like, because they're a lot more specific and less conceptual, so it wouldn't make sense to sit in on classes even if time allowed for it.

One fund manager that seems to enjoy academic-style presentations and publications is DFA, and they have a lot of well-established advisors.

    • 1
Oct 25, 2017
Comment
    • 1
Oct 28, 2017