Journal Review! (The Journal of Finance: vol 76, issue 1, Feb 2021)

I've completed the first round of this series.  I've read the first issue of the premier financial economics journal ("The Journal of Finance"), summarized its contents, assessed their value for society at large, and even took a few notes on technical concepts that came up in each.

This journal issue included 10 articles in total.  The topics covered included:

• 3 on corporate finance (articles 5, 6, 7)
• 2 on personal finance (articles 2, 4)
• 2 on information (articles 8, 10)
• 1 on the environment (article 1)
• 1 on financial products (article 3)
• and 1 on financial markets (article 9)

I rated each by the apparent social value of the authors topic, methods, and conclusions from 1 to 4 stars.  

• None were rated 4-stars, 
• 1 was rated 3-stars, 
• 3 rated 2-stars, 
• and 6 rated 1-star.

Overall the experience reading and rating these articles was at times exhausting and infuriating, but almost always stimulating and challenging.  It was like dropping into a conversation with very bright people and trying to keep up despite no one slowing down for you.

I think it's probably important to point out the high points and low points of this process. I loved Akey and Appel's paper on 'the limits of limited liability'.  They found a supreme court case that limited the about of liability companies have on their subsidiaries' pollution and showed convincingly that it led to greater rates of pollution.  It was historically fascinating, morally relevant, and their process to gather the relevant data was stimulating.

A number of other papers felt, well, bleak and disconnected.  Guren, Krishnamurthy, and McQuade wrote at length about the benefits of properly used adjustable rate mortgages (products that were subject to considerable fraud and subprime mortgages leading up to the 2008 financial crisis).  And Ai, Kiku, Li, and Tong wrote about dynamics that explain executive compensation, but assumed both that the best executives are paid the most and that their compensation substantially reduces firm performance, which sounds more like a hostage situation than a description of why these executives are 'the best'.

The rest are described below.  I encourage everyone to look at the 'one sentence summaries' in particular.  All comments are welcomed and will be moderated as needed.

Thanks for reading.  See you next month.

To a better world, through our current world,
Brandt 

1. The Limits of Limited Liability: Evidence from Industrial Pollution (link)
PAT AKEY IAN APPEL

• Notes: Supreme court case (US v Bestfoods) decreased liability of parent corporations regarding cleanup of any subsidiaries’ pollution.  Liability was limited from ownership to just operation of facilities.  Reviewed EPA data before and after decision to see if it has incentivized pollution. Concluded that they saw 17% decrease in spending to reduce polluting activities.  Compared plants in and out of places with greater parent liability, noted up to 9% in polluting without liability.
• One sentence summary: The Supreme court decided in 1998 that corporations aren’t liable for the polluting activities of their subsidiaries (unless they themselves run the factories) and, to no one’s surprise, anti-polluting spending decreased and pollution increased.
• Technical concepts: difference-in-difference (link)
• Social value: ✭✭✭. It comes as little surprise that corporate shell games enable profiting from environmental degradation, but it’s a clear problem.

2. Do Household Wealth Shocks Affect Productivity? Evidence from Innovative Workers During the Great Recession (link)
SHAI BERNSTEIN TIMOTHY MCQUADE RICHARD R. TOWNSEND

• Notes: Will the stress of having your home’s value drop (like in the housing crisis) impact your ability to do your job?  Bernstein et al say yes, that by looking at patent data (how often certain types of ‘innovative’ employees create new patents, particularly ones that are cited more often or cause higher spikes in their firm’s stock values when announced) and comparing employees of similar firms, jobs, and from similar places, they see up to a 15% decrease in economic value added from the patents created as a result of having a ‘negative shock’ to the value of your home.  The authors say this contributes to a couple debates: against the claim that financial distress in employees will cause them to work harder and add more value to their companies, and against the claim that workers carry out innovative policies for firms and are generally interchangeable.
• One sentence summary: Turns out that if your personal life starts to fall apart (like having the value of your brand new home drop during a financial crisis) that it’ll impact your work life as well (measured by how many or how good of patents some workers make).
• Technical concepts: n/a
• Social value: ✭. It’s hard to say the takeaway here, because it feels like the debate being pushed against (that workers whose home values drop may work harder) is not widespread.  Also the implications feel more frightening than anything else: should employers track more closely the financial circumstances of their employees?  Should we try harder to not have housing crises?  Or should we treat innovative employees like professional athletes that are given financial counseling to guarantee high output? The results feel obvious, but also frightening and bleak.

 

3. Mortgage Design in an Equilibrium Model of the Housing Market (link)
ADAM M. GUREN ARVIND KRISHNAMURTHY TIMOTHY J. MCQUADE

• Notes: What types of mortgages will provide better social outcomes?  Testing within an equilibrium model (like a simulation, or, as the authors put it, a laboratory setting) where multiple mortgage types (variations of fixed vs adjustable rate mortgages) can be tested out to see which results in the lowest number of defaults (as well as market volatility and homeowner income), the authors conclude that fixed rate mortgages that can be converted to adjustable rates at the moment a crisis begins provides the greatest benefit.  They do technically say that providing some kind of ‘insurance’ wherein borrowers can ‘negatively amortize’ their loans (increase their debt) when they need additional cash is the most efficient, but it may result in excess debt burden over time.  The key dynamic the authors built into their model was the negative cycle of defaults increasing housing supply, which decreases home prices, which causes more defaults; it was this dynamic that they generally argue is short circuited by having mortgages that adjust along with interest rates, as they are lowered to help save the economy in a crisis. 
• One sentence summary: adjustable rate mortgages (or the option to convert to one) can really help when rates drop, but when rates rise, well, that’s a harder question.
• Technical concepts: quantitative equilibrium model, stochastic discount factor (link)
• Social value: ✭. It’s hard to not feel like this conversation is both dated and disconnected.  I believe there is considerable value in assessing the best social outcomes for a given mortgage type, but adjustable rate mortgages, to date, have been the source of considerable fraud (e.g. rate errors) and tragedy (supposedly 90% of the subprime mortgages prior to the 2008 housing crisis had adjustable rates).  The broader implication that getting assistance to mortgage owners early during a crisis seems like an important one, but to imply that these can be done most effectively through privately-negotiated contracts that have a long and fraught track record feels far-fetched.

4. The Capitalization of Consumer Financing into Durable Goods Prices (link)
BRONSON ARGYLE TAYLOR NADAULD CHRISTOPHER PALMER RYAN PRATT

• Notes: Looking at a very large dataset of autoloans the authors control for many variables to understand the impact of the length of car loans has on the final price of used cars.  Turns out that the impact of lowering maturity of a loan (from, say, 5 years to 3 years) on average results in car buyers paying less for vehicles.  The authors contend this is likely due to greater negotiation by the buyers, as opposed to looking around at more dealerships.  They also say this speaks to the literature that says people are influenced by how much their monthly payments will be, regardless if it’s due to the interest rate or the length of a loan.
• One sentence summary: used car sales data shows that people are deeply affected by their monthly payment, regardless if it’s from interest rates or loan length, and will negotiate more regardless.
• Technical concepts: difference-in-difference, two-stage least square procedure, Oster adjustment
• Social value: ✭. Who knows.  This study shows people will fight for lower prices based on their monthly bill for big purchases (home, auto, etc.).  It isn’t clear what value this offers besides an otherwise intuitive result: people budget monthly and have to make ends meet… and a car payment has to fit into that budget as well.

5. Inalienable Customer Capital, Corporate Liquidity, and Stock Returns (link)
WINSTON WEI DOU YAN JI DAVID REIBSTEIN WEI WU

• Notes: In what ways does customer loyalty play a role in the financial well-being of firms?  Here the authors say there are two main contributors to brand loyalty (talent and brand recognition), the first one being about the people at a firm who can come and go and the second being the firm’s reputation that can’t quit or be hired and fired.  The authors build a model and use a consumer brand survey to test it.  Their study mostly confirms their belief that there’s a strong relationship between the level of talent at a firm and turnover rates, the impact of ‘financial constraints’, and returns.  
• One sentence summary: If you have a firm with a strong brand, but that brand is based on the people you employ, you’re running a much riskier operation with high returns, but also high risks keeping those people and funding their roles.
• Technical concepts: dynamic model
• Social value: ✭. This feels like an extension of a very long tradition in economics, where the structure of corporations are compared to assess the upsides and downsides of the risks they face.  It’s of a general interest, but besides cynically encouraging high talent-based companies to locate in states that enforce non-compete clauses (or to replace older ‘talent’ employees with lower paid young ‘talent’ employees), I’m not sure who specifically is helped by this.

 

6. A Dynamic Model of Optimal Creditor Dispersion (link)
HONGDA ZHONG

• Notes: Is it better for a firm to have many creditors or just one?  Is there an optimal number?  The author here builds a dynamic model to look at data over time and also looks at empirical data at a specific moment to assess what kinds of firms might have more creditors, and when that would occur.  What they see is that after performing poorly a firm is highly likely to increase the number of creditors, but better performing firms typically have more creditors. This may seem surprising on the surface, but it’s mostly due to their ability to take on more debt in total.  Lastly a few common sense findings also follow: a firm is more likely to renegotiate their debt if it means they can change the timing of when their debts are due, if a firm negotiates too much they’ll be left with less overall credit available, and if they have more cash they have a better chance of getting more creditors.
• One sentence summary: If a firm increases their number of creditors it may be because things are not going well, but it might be because things are going great.
• Technical concepts: dynamic model
• Social value: ✭. Maybe it’s important to not judge a book by its cover.  A firm with more creditors may just be performing very well and has capacity to actually take on more debt than they’re holding.  On the other hand, if they’ve taken up their capacity, it may indicate that they have been performing poorly and needed a new injection of cash.  Regardless, these studies of firm behavior are interesting on the surface, but the mind groans to identify anything beyond a curious value investor taking much use in their conclusions.

 

7. A Unified Model of Firm Dynamics with Limited Commitment and Assortative Matching (link)
HENGJIE AI DANA KIKU RUI LI JINCHENG TONG

• Notes: What explains the level of executive compensation at firms of various sizes?  The authors attempt to mix together several theories to create a big model, which they compare against data to see if it works correctly.  In the end, they claim their model shows that small and large firms do different things with executive compensation during tough times.  Small firms pay out less to stockholders and executives, and invest, resulting in faster growth.  Big firms, instead, increase executive compensation, which results in slower growth.  This is because the authors believe (according to their model) that the best executives at these big firms have more options, so they have to work hard to prevent them from leaving. A few other conclusions are reached (regarding the shape of the distribution of executive compensation, relationship between productivity changes and compensation, etc.), but this paper is mostly attempting to make a grand vision of how executives are paid given their options to leave, and the impact of the firm as a result. 
• One sentence summary: If we live in a world where big firms sacrifice growth to keep their executives (who they perceive are the best) and small firms have great growth because they don’t pay theirs (who they perceive aren’t as great), then what exactly makes an executive great?
• Technical concepts: assortative matching (link), limited commitment (link), stylized fact (link), Gibrat’s law (link), power law (link)
• Social value: ✭. This paper is less concerned about social problems in the economy.  it is simply concerned with finding a way to explain how the world works as is.  It seeks to justify how we have reached a point in which executives at big companies receive seemingly outrageous compensation levels by relying upon theories that say those executives, in the labor market, are perceived to be the best.  The implications they provide are practically non-existent and the authors don’t attempt to use their study’s model for anything outside mirroring our world.  In doing so it helps no one, but it does depict a universe whose assumptions and conclusions (the best executives need to get paid, if they don’t get paid they leave, but when they get paid their company grows less quickly) show our society as one in which those at the top of the corporate ladder hold the rest of us at a ransom that we can only justify with irony.

8. Information Consumption and Asset Pricing (link)
AZI BEN‐REPHAEL BRUCE I. CARLIN ZHI DA RYAN D. ISRAELSEN

• Notes: How exactly do financial news stories impact the values of firms?  Traditionally, it’s been expected that if a big news story came out about a specific firm or industry, one could expect to see these firms’ stock prices react.  The authors here, however, decide to consider how news can spillover and impact unexpected firms not directly mentioned in news stories.  Here, by gathering data on stock prices impacted by previous news stories and on the ‘consumption of data’ (or when stories cause firms to be read about or searched for online), they create a model to predict how much the ‘other’ firms will be impacted by news stories that tend to spillover and change their stock prices.  The authors check for other explanations beyond impacted ‘information consumption’ (firms in similar industries, firms that tend to be traded together, etc.) and even other explanations (temporary price changes, price changes due to previous mispricings, etc.), but none are as convincing.
• One sentence summary: the news plays a role in how investors price (how much they’ll pay to buy and sell) equities even for equities that are not directly referenced in the news stories, and data showing when investors read about certain stocks is helpful in predicting price changes.
• Technical concepts: panel regressions, monotonic, risk premium
• Social value: ✭✭. This feels like an extension of the interesting discussion around not just the efficiency of stock prices (how accurate are they), but also around the power of the news, both in how it’s announced and when it is read.  I was more interested than anything to find out that places like Google and Bloomberg are tracking this ‘information consumption’ data, because it seems to show the human side to the market and role of the media in directing our investment choices.  The authors’ implications are meagre and constrained to simply pointing to their empirical results, which is unfortunate, because if these types of studies could be directed to compare sources of information (like with Reddit and the Gamestop stock rally) we may start to better understand who exactly is moving the markets the most, when, and how.

9. Learning From Disagreement in the U.S. Treasury Bond Market (link)
MARCO GIACOLETTI KRISTOFFER T. LAURSEN KENNETH J. SINGLETON

• Notes: What can help us predict future interest rates and yields in the bond market?  Traditionally, basic fundamentals were considered in making these predictions (output, inflation, etc.).  However, the authors try a Bayesian model wherein predictions consider ‘disagreements’ based on market surveys, and incorporate updated levels of these disagreements as new information arises.  The authors show that this process is actually more accurate (they claim 50% more accurate) and the information held within disagreements is likely connected to uncertainty around future government fiscal policy (particularly around post-recessionary periods). 
• One sentence summary: Predicting bond yields and interest rates can be very hard, but how much market participants disagree can be factored into predictive models and in many cases (particularly after a recession) are much more accurate.
• Technical concepts: Bayes theorem, risk premium, dynamic term structure model, Root-mean-square deviation
• Social value: ✭✭. Here the primary takeaway doesn’t seem to be the ability to make better bond-yield predictive models.  The main takeaway seems to be more connected to using disagreements as an input toward financial predictions.  When we disagree, it says something about what may happen in the future, at a significant level.  Since, in this paper, this takeaway is most significant in the weeks and months directly after a recession, I think the authors are right to say this has something to do with uncertainty around governmental fiscal policy.  Those are periods of time in which debates rage about when the government should end its interventions in the market.  Those debates probably play a bigger role in Federal Reserve and US government interventions, than we realize.  Though they try to be objective, they cannot entirely remove themselves from the world around us.  

 

10. Information Inertia (link)
PHILIPP K. ILLEDITSCH JAYANT V. GANGULI SCOTT CONDIE

• Notes: If stock prices supposedly reflect all available information, then why do we see prices trend toward new values over time after major announcements (earnings, inflation, market news, etc.)?  The authors here create sophisticated models toward an argument that, they believe, shows that because of investor’s aversion to risk and ambiguity (not knowing the relationship between the available information they have and how it will impact their stock prices) they don’t move on information until they see prices start to move.  They also believe that if the stock value is riskier, more ambiguous, and the information announcement is more broad (not just about one company), this shift in price reflecting new information will be more dramatic.  The fact that this ambiguity plays such a big role, and even with publicly held data, are the key findings.
• One sentence summary: Stock prices don’t always perfectly reflect all available information, particularly for risk-averse investors who wait to see where the market is moving before acting, and, more importantly, in situations where the relationship between the factors they are analyzing are ambiguously related to the stock price.
• Technical concepts: CRRA/CARA, joint normal distribution, Sharpe ratio
• Social value: ✭✭.  By tying this back to the actions of less educated, less financially literate, and less wealthy households, the authors extend these conclusions in an interesting and valuable way.  These market participants are afraid.  They are afraid, in general, of taking major risks and so they trade less often and, most likely, rely upon passive investment techniques.  The authors state that their delayed response is due to ambiguity, or simply not knowing how the factors these investors use to determine what to invest in and at what price impact the eventual value of their stocks.  Where does this ambiguity arise?  It’s not entirely clear, but one can infer for the reasons spelled out above (less sophisticated investors have greater ambiguity).  It is their uncertainty that prevents them from acting immediately and results in those who do act to pick up a nice profit, which only becomes more pronounced as the ambiguity grows and the announcement is of greater value.  What are the implications?  These are not fleshed out, but one hopes that it becomes obvious that markets are not just composed of bears and bulls.  There are many sheep present as well, and their actions play a big role in how fast news is understood and stocks are priced.