Should economic policy-making be left to economists alone and should the entire economic management structure in the U.S be left to institutions without a political interest?
In our book, The Role of Policymakers in Business Cycle Fluctuations, we do address this matter as it pertains to monetary policy.
While the existing research does suggest more politically independent central banks (i.e., monetary authorities/policymakers) tend to be correlated with lower inflation, there are still some significant questions about the proper role of policymakers.
If the goal is less inflation and output volatility, then we argue a policymaker’s role is narrower and should be focused on inflation stability first.
This is also a plea for modesty. We expect too much from policymakers because the models and information shortcomings they use and face leave a good deal of room for improvement.
In the interim and as a second best practice, the long-running argument for policy rules, say an inflation rule, appears to provide a useful and productive role for policymakers.
Underlying this plea for policy rules is the assumption that the economic process involving prices, profit and loss, voluntary exchange, and creative destruction dominate day to day economic activity.
It is a bottom-up process. Policymakers — who sit at the top — in this environment are not seen as “steering” the economy but taking a secondary role that assists in this dynamic and evolutionary process.
In your article “Modeling and testing the diffusion of expectations” you explain the boomerang effect’s impact on economic forecasts and how it contributes to economic instability. Will the economy become more stable with more sophisticated predictive modeling, and does a viable economic model exist that is more stable than the American boom-bust economy?
In that paper we turn around the usual assumption (in monetary theory) that the better informed agents “lead” the less informed agents.
Here we assume a world where policymakers not only lead the public but also depend on an accurate reading of public reaction to policy. Specifically, we model what happens when better informed agents (i.e., monetary policymakers) have flawed information on the less informed agent’s information?
We find that policy success depends crucially on how policymakers interpret that information. The unintended consequences are real.
The issue then is whether policymakers will double down and try to correct their prior mistakes and, in doing so, make new mistakes and destabilize things further.
That is one complaint made about policies in the 1970s. Policymakers were late in realizing just how much the public’s inflation expectations had changed and policy errors compounded themselves.
If the term “viable” model is interpreted to mean one that produces accurate forecasts, then no viable model exists.
Part of the problem is the modeling tradition. What has occurred in the past decades is the great facility with mathematics and statistics has lead to placing an emphasis on truly ingenious changes in technique.
Yet, the mathematics and modeling changes we employ are limited in that they are not capturing behavioral and economic fundamentals. This limitation is particularly true as it pertains to capturing things such as creative destruction and Adam Smith’s contribution on the ever changing dynamics of specialization and trade.
Having said this, we should not make perfection the enemy of the good. We are a huge supporter of modeling and it needs to continue.
Indeed, the shortcomings noted above led to the creation of the National Science Foundation’s (NSF) Empirical Implications of Theoretical Models (EITM) initiative. The future emphasis will be on getting data that accurately capture the dynamics noted above but also using models and tests that are transparently linked in their parameters.
Why did you gravitate toward studying public policy and econometrics, and how did that interest develop over time?
The primary motivation was the stagflation of the 1970s. People were having trouble economically — losing their jobs, their incomes not keeping up with inflation — and it became important to understand what the causes were. Understanding policy was the key to figuring this out, and formal modeling and econometrics (statistics) were the primary tools that academics were using. So, you needed to become an informed user of those tools as a first step in the overall process of understanding policy and, more importantly, evaluating the potential success of policies.
What would the layman find most interesting and accessible within your research?
We have published some less technical papers on both monetary policy and research methodology (EITM). The real-world implications of this work manifests itself in two ways.
In the case of policymakers — monetary policymakers — the policy prescriptions are intended to assist in creating an environment where uncertainty for businesses is reduced. Businesses struggle to anticipate consumer demands and the onslaught of new competition. In our way, by supporting inflation stabilization, the relevant economic actors (i.e., businesses) are more likely to get a more accurate assessment of changing price (demand) signals and, in so doing, enable them to adapt in ways that sustain their enterprises.
For research methodology, the ultimate goal of EITM is to get to the truth on how things relate to each other. We believe when you have models transparently related to tests that feedback between the two supports an evolutionary trial and error process that allows us to get a better handle on what we are trying to understand. Now, the “truth” is hard to come by and models and tests, even when they are transparently linked, can lead to a dead-end with subsequent research. But, the point is there is a necessary coherence that assists this stream of research that builds directly on the prior work.
What is the most interesting story in your academic career?
The greatest adventure is helping build the Hobby School of Public Affairs at the University of Houston. We started as a very small scale policy center 10 years ago and have now evolved to a policy school. If you talk to people in the business world it is akin to a ‘start-up’ with all the attendant risks, frustrations, uncertainty, and rewards one would expect. One lesson in this whole “birthing” process at the Hobby School — and a key to overall success — is to not be afraid to fail and to even risk embarrassment and ridicule along the way.
There have been many fortuitous incidents. They are not so much personal events or milestones in research, etc., but in the people you meet, individuals who indirectly influence your research, etc. Who does not have a long list? No exception here. Many people have had a big and positive impact. To name a few: John Aldrich, who was the prime force and mentor (and EITM “instigator”) in graduate school at Duke.
That influence endures. Frank Scioli — who made the NSF experience unforgettably positive. Among other things, he helped institute the EITM reforms but he was a total pro in all ways. Boy, can he make you laugh too. Another was Governor Bill Hobby. He gives you a vantage point that not only encompasses decades of public service but it is also combined with his experiences in the private sector and academia. He is an enormous force for good.
With all the funding in the world, what would you have researched and how?
Data, data, data, and more data. This is definitely not an original thought. Data has improved enormously in the past 50 years. But, great data — particularly data that captures key elements of human behavior and the economic dynamics noted above — are still a work in progress. Panel data would be at the top of the list. These data are big ticket items but you would get what you pay for.
Another funding target is providing stipends for graduate students and faculty to subsidize technical training. NSF has done this in the past, but with unlimited funding this would be something to scale up in a big way. For most social scientists, graduate school simply does not afford the necessary time to become truly competent in an array of methodological tools. Methodological training is too siloed. Moreover, this should be seen as a process of life-long learning. It never ends.
And one is struck by the fact that the best among us have this self-improving mindset. People like Chris Achen and Lin Ostrom transformed themselves (methodologically) from what they did earlier in their careers to the benefit of the rest of us.