In Memory of Bob Rossana

Robert “Bob” Rossana died back in February of this year. I was inadvertently left off of the university announcement email and did not hear about it until much later. Bob was my dissertation advisor. He meant a lot to me and so I want to share some thoughts.

In graduate school, there were four macroeconomics courses. Two core classes and two field classes. Bob was instrumental in getting this four-course system started, as well as a “macro lunch,” in which any graduate student interested in macro was required to attend and present.

Many of my fellow classmates found him intimidating. I cannot say for certain why other students felt this way, but I suspect that it was Bob’s high standards. He expected a lot of us, but he was always fair. Above all, he wanted us to push the boundaries of our understanding.

Those of us who chose to write in macro got to know a different side of Bob. Visits to his office involved not only serious discussions about research and economics, more generally, but current events and funny stories. He had a great sense of humor.

He loved to talk about current events and filter them through the lens of economics. During these conversations, he frequently emphasized to me that the ability to think like an economist was a powerful tool and that one should never let economics become inconvenient. What he meant was that one should never use his or her position as an economist to advocate a particular ideological position. Any sort of policy advocacy should be based on a rigorous application of economics.

In these meetings, I also learned that he had the same high standards for what he deemed good research as he did for his students. We had many conversations about what constituted good, serious scholarship. As I have gained my own experience in the field, I have also developed strong opinions about what constitutes good research. I recognize now that a lot of the opinions I have today were shaped by these discussions.

He did not believe in a distinction between theory and applied work. Applied work should be a direct test of a theory. This is reflected in his underappreciated paper that demonstrates how to test hypotheses for a system of equations. He also frequently emphasized that one must be careful about trying to explain the data-generating process when the data are often sensitive to the nature of measurement. This was evident in his own work as well. For example, Long and Plosser had found that most macroeconomic time series followed a similar time series process. This evidence was used early on as support for real business cycle theory. Bob, along with his friend and co-author John Seater, showed that this Long and Plosser result was predicated on the use of annual data. The last paper he was working on (to my knowledge) was a unique test of money neutrality. I don’t see a recent copy online, but hopefully his co-author will see it through to publication. He was a serious scholar. A relatively recent version of his CV is here.

Bob was very business-like as a professor and an advisor. Things were either done well or not. I remember when I presented him with the first complete draft of my dissertation. I naively assumed that a completed draft meant that I was basically done. He called me to his office and listed entire sections of the dissertation that needed to be re-written – most sections, if I’m being honest. I was a bit dejected, but upon reflection, he was more than kind. When I got my first revise and resubmit during my final year of grad school, he congratulated me on my progress, but told me I should’ve sent it to a better journal. (He was probably correct about that as well.)

I had so much respect for him. Even after I had defended my dissertation, I would often address him as “Dr. Rossana.” One day, he sent me a quick email that said “Josh, you need to call me Bob. We are colleagues now.”

It was his business-like nature that made praise all the more meaningful. The last time I saw him in person was when I came back to give a seminar. I was waiting on one last signature for tenure. After my seminar, we sat in his office for a couple of hours. We talked about my job and my all-but-certain tenure case. He asked me about my wife and kids and how they liked Mississippi. When I mentioned the long road trips to visit family, he shared memories of car trips with his sons. Before I left, he said, “I want you to know that we’re proud of you. You’ve done well.” That meant a lot to me. More than he probably realized.

He died in February. Cancer. I never got to say goodbye. Or one last thank you. So I wrote this as that final goodbye and that final thank you. Hopefully it reaches him.

The Coronavirus and Lessons for Preparedness

My new Mercatus policy brief is up. It is entitled, “The Coronavirus and Lessons for Preparedness.” You can find it here.

Podcast on Maritime Policy

I recently recorded an episode of The Economics Detective podcast with Garrett Petersen. The subject of the podcast is my paper, “U.S. Maritime Policy and Economic Efficiency.” A link to the paper is here. A link to a blog post I recently wrote about the paper is here.

On Maritime Policy

Some time ago, I was listening to NPR and the hosts were discussing the Jones Act. For those who are unaware, the Jones Act requires that any shipping done from one U.S. port to another U.S. port must be carried by a U.S.-flagged ship with a U.S. crew that was built in the U.S. The hosts claimed that they could not find an economist who could explain why this sort of law existed. The policy seemed like a run-of-the-mill protectionist giveaway. My reaction was much different.

Here we have this law that is nearly 100 years old and no economist can explain why it exists? The only plausible explanation is that this 100-year-old law is a protectionist giveaway? This seems unlikely. One would have to believe that democracy is incredibly inefficient. This is especially true when one hears the estimates of the costs of the Jones Act. Why are voters in the U.S. just leaving billions of dollars on the sidewalk?

There is an entire literature on elections and accountability that seems to suggest that elections discipline the behavior of politicians to follow the will of the voters. However, as part of this literature, this can lead politicians to give greater weight to special interest groups. These groups tend to have more information than the average voter and provide more feedback to politicians. As a result, politicians might be too responsive to special interest groups in trying to achieve re-election. Of course, what seems like an obvious question (to me) is how special interests could convince politicians to support this legislation for so long without some other special interest group emerging to get the inefficient policy repealed. In fact, I have made the argument (along with my co-authors, Brian Albrecht and Alex Salter) that, in a world with endogenous entry of special interest groups, long-lasting policies must either have modest costs or actually efficient.

Given my prior beliefs on the endogenous entry of special interest groups, I figured that something like the Jones Act must either be substantially less costly than conventionally claimed. I decided to study this more closely and what I discovered is that there is a clear and consistent pattern of subsidizing the shipping industry throughout U.S. history. In addition, there is also an argument that such subsidies are actually efficient. So, I decided to write a paper about it. For those who don’t want to read the whole paper, what follows is a synopsis of my argument.

Consider the following basic idea. During times of peace, a navy needs less capacity than during times of war. At the same time, a navy needs to have the ability to expand rapidly when war occurs. Excess capacity during times of peace could be wasteful. A potential solution to this problem is for the navy to use private ships to assist them during times of war. Of course, there are several problems with doing so. First, diverting private ships for naval use might impair international trade. Second, foreign crews might not want to have any part in another country’s conflict. Third, if the government offers to pay ships for their services once conflict has begun or is imminent, it might be subject to a hold-up problem. Fourth, if the government simply expropriates the ships for naval use, the expectation of expropriation will cause the shipping industry to underinvest.

Given the need for a naval auxiliary, it would seem like some sort of Coaseian bargain is in order. One possibility is for the government to provide peacetime subsidies to the shipping industry in exchange for this industry’s services during times of war. This sort of thing is not without precedent.

What we have here is a possible argument for why the U.S. government offers maritime subsidies. But how would we know if this is actually the case? How do we know that I haven’t crafted a just-so argument to put a thumb in the nose of those who say there’s no justification for policies like the Jones Act? Well, I would suggest that we would observe multiple things. First, we should observe a clear and consistent pattern of maritime subsidies across U.S. history. Second, we should observe similar subsidies for other transportation services.

It turns out that both observations are true. As I detail in my paper, there is a clear and consistent pattern of subsidizing shipping and revising and updating these maritime subsidies in the run-up to military conflict or in the aftermath of conflict. Furthermore, the legislation for these subsidies, as well as speeches by FDR and Richard Nixon, explicitly outline the importance of these subsidies for the role of the merchant marine as a naval auxiliary.

There is also a history of doing this with other forms of transportation. In the decade prior to World War I, the U.S. became concerned about the number horses that would be available during times of war. Again, the U.S. could potentially purchase horses from private citizens in the event of war. However, military horses do not always have the same characteristics as the horses used on farms. Furthermore, mother nature determines the “time-to-build” for horses such that no reallocation of resources or productivity improvements allow military-ready horses to be produced any faster. As a result, the U.S. government set up remount depots and subsidized horse breeding.

Similarly, during both World War II and the Korean War, the U.S. government requisitioned private planes to provide airlift during the war. The role of aircrafts was later formalized with the creation of the Civil Reserve Air Fleet in 1952. This program allows the government to use participating member planes for airlift during times of war. In exchange, the government uses the participating airlines for airlift during times of peace and the amount of business and compensation provided is a function of the complementarity of the planes for military airlift.

In other words, what we see is a clear and consistent pattern of Coaseian bargains across time and different methods of transport as it relates to creating additional transportation capacity during times of war.

This finally brings me back to the Jones Act. As I detail in my paper, the Jones Act is part of a larger subsidization effort on the part of the U.S. government since its founding. In fact, the cabotage restrictions that are so notoriously criticized actually originated in 1817.

As I stated at the beginning of this post, much of the criticism of the Jones Act is based on its perceived astronomical costs. If one takes these costs as given, then when can understand why it is so hard to find an economist who can explain why it exists. However, what I have found is that these costs are dramatically overblown because they fail to consider the proper counterfactual. When estimating the costs of the Jones Act, most critics compare the cost of U.S.-flagged ships engaged in U.S.-based trade to the costs of foreign-flagged ships engaged in international trade. When you do so, what you find is that the cost of U.S. ships is almost 3 times as high as the foreign-flagged ships. This makes the Jones Act seem really expensive! Shouldn’t we just let the foreign-flagged ships do the work?

What this comparison ignores is that these foreign-flagged ships often fly so-called flags of convenience. These flags allow them to avoid the sort of taxes and regulations that they would face by flying the flag of a major developed country. Port-to-port U.S. shipping is considered domestic production. As a result, any ship engaged in U.S. port-to-port shipping would be subject to U.S. tax law, immigration law, labor law, and environmental laws. Perhaps most important among these compliance issues is immigration law. Under current U.S. law, foreign crews could not work on a ship conducting U.S. port-to-port shipping. This means that the labor costs of the crews on foreign-flagged ships would be substantially higher. In fact, back-of-the-envelope calculations suggest that once labor costs are taken into account, the cost of U.S. ships would only be 8% higher than foreign-flagged ships rather than nearly 3 times higher – and this is without taking into account taxes and other regulatory costs.

So, perhaps the reason that voters in the U.S. aren’t reaching down to grab billions of dollar bills on the sidewalk is that they recognize these dollars as a mirage.

None of this is to say that current maritime policy is optimal. In fact, I detail criticisms of current policy in my paper and offer suggestions for reform. Nonetheless, my paper illustrates that there is a compelling reason that countries offer maritime subsidies. This significantly changes the relevant counterfactual. So there is at least one economist who can now explain the existence of the Jones Act to the good people at NPR.

Macro Musings

This week I was a guest on David Beckworth’s Macro Musings podcast. We discussed my policy brief on the labor standard as well as monetary policy more generally. Here is a link for those interested.

Updates

A couple of updates:

• The topic of this month’s Cato Unbound is J.P. Koning’s proposal for the U.S. to issue a large denomination “supernote” and to tax that note as a way of punishing illegal activity. I will be contributing to the discussion this month along with James McAndrews and Will Luther. You can read J.P.’s lead essay here. The response essays will be linked below the lead essay. My response essay will appear next week.
• My paper with Alex Salter and Brian Albrecht entitled “Preventing Plunder: Military Technology, Capital Accumulation, and Economic Growth” has been accepted at the Journal of Macroeconomics. I think that this paper is based on a really interesting idea (biased, I know). The basic idea is that military technology is a limiting factor for economic growth. We also suggest that both economic growth (at least to some degree) and state capacity could be driven by this common factor.

Monetary Policy as a Jobs Guarantee

Today, the Mercatus Center published my policy brief on the idea of a “labor standard” for monetary policy that was first proposed by Earl Thompson and David Glasner.

Towards an Alchian-type Approach to Political Economy

In my previous post, I discussed what I called the sleight of hand of an Olson-approach to political economy. The basic idea of that post was that Olson’s theory of concentrated benefits and dispersed costs is often used to malign policies deemed to be inefficient. The sleight of hand aspect is as follows. First, the economist deems a particular policy to be inefficient using a standard theoretical model. Second, the economist hypothesizes that the reason we have such an inefficient policy is due to special interests getting what they want because the costs are dispersed. Third, the economist examines either in historical detail or through regression analysis the role of special interests in getting the policy implemented. Fourth, if special interests are found to have had an effect on the policy being put into place, the economist concludes that the reason we have this inefficient policy is due to special interests. However, the inefficiency of a particular policy is determined by some theoretical model. The empirical finding that special interests had a marginal effect on the policy’s implementation is then used to explain why we got this inefficient policy. Whether or not this is the correct interpretation of the empirical result depends critically on whether the theoretical assertion is true!

Let me elaborate on this point using an example. Consider the example of pollution, which is a principles of microeconomics textbook version of an externality. Suppose that in the absence of special interests, such as environmental groups, pollution would go untaxed. Empirical evidence would show that a tax on pollution was due to the influence of special interest groups (the environmental groups). If we had no concept of externalities and we used the perfectly competitive model as our benchmark model, the conclusion would be that special interests were to blame for this inefficient policy. However, since this is a commonly understood externality, one would not conclude that special interests were to blame for an inefficient policy. On the contrary, the special interest groups would be the reason for implementing the socially optimal policy. In other words, finding evidence of the role of special interest groups does not tell us anything about what is efficient or what is optimal. To do so, we need an explicit theoretical argument or model. Not only that, to come to the correct conclusion we need a correct theoretical model in the sense that it addresses relevant factors, such as externalities.

In my previous post, I criticized economists for too often simply asserting that a policy is inefficient and subsequently applying Olson’s model to explain why we get such stupid, inefficient policies. I also argued that political economy should shift to using an evolutionary argument. In that post, I was short on some of these details. As a result, in this post I want to outline what I meant by this.

In 1950, Armen Alchian published a paper entiled “Uncertainty, Evolution, and Economic Theory.” In that paper he outlines an evolutionary approach to economic theory. Specifically, he argues that it might be misleading to describe firms as profit-maximizers. The reason is that when firms face decisions, there might be a distribution of outcomes across each decision. If these distributions overlap, then it doesn’t make sense to think of the firm as maximizing anything. For example, one distribution might have a higher average profit, but also be associated with a greater variance in profit than some other possible decision. So if firms aren’t maximizing profit, what are they doing?

Alchian suggested that we think of firms as practicing trial-and-error and imitation. Firms try certain things to see what works and imitate things that worked for other firms. Along the way, some firms benefit from good luck and other firms suffer from bad luck. Nonetheless, through this process of trial-and-error, imitation, and uncertainty, the profit mechanism ultimately determines what firms are able to stay in the market and what firms must leave the market. Firms that earn a profit are able to continue operating. Firms that are losing money will be forced to drop out of the market. The profit and loss mechanism therefore selects for firms who are making a profit. This might be due to the firm’s decision-making or it might be due to luck (or some combination of the two). Nonetheless, the economist should be able to explain the success (or lack thereof) of firms. An economist can look at the characteristics and decisions shared by the surviving firms and contrast those characteristics and decisions of the firms that have left the market. In doing so, one can get the sense of the role of decision-making and luck as well as the types of decisions that have proven successful and unsuccessful.

In my view, political economy would benefit from the same sort of approach. Rather than start with a baseline model to determine whether some policy is efficient or not, one should start with the policy itself.

1. What was the primary justification for the policy? More importantly, under what grounds could we consider the policy to be efficient? These questions help to set a much more relevant benchmark than some abstract model of the economist’s choosing.

2. Once these questions have been answered, the economist has some general idea about the conditions under which the policy would be considered efficient. Now, one can take an evolutionary approach to the policy. Did the policy survive for a long time and/or is it still around? What other states or countries have adopted the same or similar policies? How did states and/or countries adopting the policy perform along the relevant dimension in comparison to others? Did any other states/countries have similar policies and abandon them? What happened if they did?

The answers to these questions help to determine the conditions under which the policy survived and the relative success of those places that implemented the policy. This can help to determine if the policy actually achieved what it was supposed to and/or whether the policy is consistent with conditions under which it could be considered efficient. In addition, if the policy seems to have been an efficient response to a particular problem, it is then possible to examine why some places got rid of it and had to bear the cost of doing so thereafter.

In short, I think that it would be useful to do something in political economy with respect to government policy today that Pete Leeson has done with policies and institutions of an earlier time and place. Leeson’s work often starts by examining some policy, law, or institution that seems completely weird, strange, or backwards. He then starts with the premise that it must have been efficient. He outlines the conditions under which the policy or institution would have been efficient and then tests that theory by examining what would have to be true for his efficiency hypothesis to be correct. This approach to political economy or public choice is clearly enlightening, judging by Leeson’s publication record. However, I notice a reticence on the part of many political economists and public choice economists to take the same approach to more recent policies and institutions. The attitude toward more modern policies and institutions seems to be that we “know” that policy X or institution Y is inefficient because economic theory tells us so. Therefore we need to explain why it exists. But how do we “know” this any more than we “know” that trial by battle was a backwards and barbaric practice of no practical use? Leeson’s work showing that trial by battle was a good way of eliciting the true value that particular claimants placed on land was actually quite efficient. So maybe we should be a bit more humble about modern policy as well.

The Sleight-of-Hand of Olson-esque Public Choice

“Long-surviving democracies could therefore hardly have been dominated by the charlatans, simpletons or crooks that economists typically portray in characterizing democratic representatives.” — Thompson and Hickson (2000)

The early days of public economics (at least as a distinct field) were essentially normative. The basic idea was that economists could use economic theory to examine market failures and devise policies that correct for those failures. A quintessential example is the case of externalities. Suppose, for example, that a particular type of production produces pollution. This cost is not limited to the firms or those working for firms. This cost affects (potentially) all members of society. The social cost of production is therefore greater than the private cost. Since firms are unlikely to internalize this social cost, they tend to produce “too much.” In order to correct for this, the economist would likely recommend taxing production at a sufficiently high level to reduce production to the socially optimum level (i.e. the level that takes into consideration both the entire cost). In this world, the economist largely plays the role of technocrat, identifying market failures and offering corrective policies.

James Buchanan, one of the pioneers of what is now called public choice theory, suggested that public economics should take a different approach. In particular, he suggested that public economics should concern itself with positive economics (i.e., an analysis of “what is” rather than “what should be”). The field of public choice recognized that the process through which policy is enacted requires the deliberate actions of politicians and other policymakers as well as voters and special interest groups.

One view of policymaking that emerged in the public choice literature is most closely identified with Mancur Olson. According to this view, the democratic process contains policymakers who are capable of supplying policy and the general public who have demands on policy. The general public will tend to form coalitions, which we might call special interest groups. As the name implies, these groups are organized around a common interest. These interest groups then go to politicians and other policymakers with their concerns and petition for policies that provide direct benefits to their group. This process tends to be effective for special interest groups because politicians are self-interested. A politician cares about getting re-elected. As such, the politician has an incentive to give the special interest groups want they want because the benefits are concentrated within the group, but the costs are dispersed throughout society. Since the costs are so small for the average voter and/or taxpayer, the marginal cost of taking action to oppose the policy exceeds the marginal benefit.

The usefulness of this theory is that it enables economists to explain the existence and persistence of inefficient policies. One shouldn’t be surprised to observe inefficient policies if those policies are providing concentrated benefits and dispersed costs. This is no doubt an insightful positive approach to the behavior of politicians and the process of policy determination.

Despite the theory’s usefulness in explaining “what is”, empirical applications of this theory often engage in a sleight-of-hand technique. For example, some public choice scholars studying a particular policy will start their analysis by using economic theory to examine a particular policy. If the policy is found to be inefficient in theory, it is natural to ask why the policy exists. One hypothesis is that this is simply an example of Olson’s theory. The public choice economist can then go back and analyze who benefits from the policy and what politician(s) supported the policy to see whether it is a simple application of concentrated benefits and dispersed costs. In the case of simple legislation, the public choice economist might even estimate a regression model of the likelihood that a particular legislator voted in favor of the policy using data on special interest influence.

The reason that I say that this application of Olson’s theory is a sleight-of-hand is as follows. This sort of analysis starts with the idea that the policy is inefficient and then empirically examines whether this is a case of special interest influence. But this is an empirical test used to justify a theoretical conclusion. In other words, economists identify a theoretical inefficiency and determine empirically that the reason the policy exists is due to the influence of special interests.

Why does this matter?

First, this approach to public choice is making a similar sort of mistake that early public economics was making. The economist starts with a theoretical model and analyzes the policy within the context of the model. If the policy is inefficient in the model, then the economist is left to explain why such an inefficient policy exists. But what if the model is wrong? What if the policy is correcting for an inefficiency that the economist ignored?

Second, this sort of empirical evidence can never tell us whether or not the policy is inefficient. In fact, it would be surprising if one didn’t find evidence of special interest influence on legislation. This is because special interests will promote their own interests, regardless of whether the policy is welfare-improving. So an economist will be likely to observe special interest influence both in cases when the special interests bring an inefficiency to the attention of policymakers and when they are just looking for a giveaway.

For a proper analysis of this Olson-based approach, economists need to develop a much better understanding of the black-box nature of the political process. The idea that special interests shape policy is not surprising. However, different political systems will select for particular policies. Countries with a given set of institutions might select for policies that appear to be inefficient, but in reality are efficient responses to some social problem previously ignored by economists. Other institutional structures might select for giveaways to special interests.

For the Olson-based approach to be useful for evaluating the efficiency properties of policy requires an evolutionary approach. Thus, the analysis of policy requires returning to the point at which the policy was adopted and attempting to identify what inefficiency the policy could possibly be aimed at eliminating. Did the policy persist and for how long? Then, one can look for whether other places adopted similar policies. If so, did the policy persist and for how long? Does the policy seem to have eliminated these inefficiencies? Among the places that didn’t adopt the policy, did they turn out better or worse in regard to this supposed inefficiency? Why did some places adopt the policies and not others? What are the institutional differences that explain what policies were selected for?

The answers to these questions seem substantially more important than the results of some probit regressions of yay or nay votes.

In Memory of John Murray

This post is a bit different than normal. Most of my posts are about the minutiae of economic theory or controversies. Today’s post is personal. All of us in academia have a number of important people who have helped us in our intellectual journey and career. I have been fortunate enough to have a number of such people in my short career. One of these people was John Murray.

I first met John as an undergraduate. At the time I was a history major, but I had started to take an interest in economics. I went to see John in his office, he being the undergraduate adviser at that time. We had not met before that day. In fact, I could not have taken more than 3 courses in economics at that time. I remember that he was delighted that I was interested in economics given that I was currently a history major. After all, John was an economic historian. When I took John’s course in economic history, it really opened my eyes to thinking about history in a completely different way. I could tell, even then, that he really enjoyed the unique perspective that economists brought to the study of history. He also had a dry sense of humor that I enjoyed.

John and I always kept in touch. When I accepted my job at the University of Mississippi, John had just accepted his position as the J. R. Hyde III Professor of Political Economy at Rhodes College. He thought that it was funny that we had accepted jobs just an hour drive from another. After we moved, I invited John down here to give seminars and he invited me up there to give a seminar. The last few years, he made a habit of coming down to Oxford to have lunch with me about twice a year. While these were technically lunches, we would often chat for a couple of hours when he would visit. We would talk about our kids and our research and he would always give me advice. We would also talk about the craziest economic theories that we thought might be correct. He also loved that I had recently taken an interest in applying modern macro to historical events. At our last lunch together in October, he was especially excited to talk about how I’d recently been given tenure and how much I had accomplished from that day I initially met him in his office.

I say this was our last lunch because John died last Tuesday. He was 58 years old.

John was not only a great mentor, but an excellent scholar. John’s work on the communes of the religious group known as the Shakers is an important work on the role of incentives in the context of particular institutional environments and should be a staple of law and economics courses. He also wrote a fascinating book on an early form of health insurance in the U.S., industrial sickness funds. He was also on the editorial board of the Journal of Economic History and Explorations in Economic History. His Google Scholar page is here. I imagine that all scholars want their work to be remembered fondly. So hopefully readers will pursue some of these links.

John was one of the most genuinely nice people that you could ever meet and he had a great laugh. He was the first person in his family to go to college and once upon a time he was a high school math teacher. Having grown up in Cincinnati, he was also a Reds fan, which I am proud to say I never held against him. His Rhodes webpage has a really great description of his life and his research in his own words, which can be found here.

The last interaction I had with John was through email a couple of weeks ago. I had sent him one of my latest papers and he told me that he was really excited to read it and discuss it. Our next lunch would have been planned for the next month or so. I wish now that I’d scheduled it sooner. I miss those lunches already. All I can hope is that John is somewhere saving me a seat for our next lunch.