By William K. Black
April 19, 2016 Bloomington, MN
The journalist Adam Davidson has written an interesting article about economics and Bernie Sanders. As an economic adviser to Bernie I found his take on Keynesian and institutional economics of considerable interest. Institutional economics, contrary to Davidson’s take on it, is thriving and the University of Missouri at Kansas City has long been a center of institutional economics. (I am one of the scholars at UMKC that works largely in this field.) Davidson treats institutional economics, which overwhelmingly studies microeconomics and the “micro foundations” of the economy as having been rendered obsolete by the transformation that Keynes’ insights sparked in the study of macroeconomics.
[Institutionalists] played significant roles in government and academia into the 1950s. John Kenneth Galbraith was one of the most (and last) prominent members of the school.
By then, economics had been entirely transformed by the work of John Maynard Keynes and his 1936 masterwork, “The General Theory of Employment, Interest and Money.” The book more or less invented the field of macroeconomics — the study of how an economy works in the aggregate. And in the decades since, most academic economics and nearly all of government economic policy has revolved around Keynes’s insights.
Davidson’s thesis is unsupportable as a matter of logic, history, and macroeconomic theory.
Logically, it is not possible that changes in macroeconomics introduced by Keynes made microeconomics irrelevant. Ecological studies are vital, but they do not make botany and zoology any less important.
Historically, Davidson’s claim is incorrect. Institutional thought by classical economists began centuries before Veblen. Some of Adam Smith’s most famous points are institutional. He warned against allowing corporations, emphasizing what he saw were severe “agency” problems inherent in the structure. He warned that the most innocent of institutions, by bringing the owners of competing firms together, were an aid to the creation of cartels – and warned that it was impossible and inappropriate to attempt to ban such meetings. Smith explained why a common institution of his day – small merchants making goods (bread and cuts of meat) whose quality the average consumer could determine – would create a strong incentive to build a reputation for good quality, particularly in a village or township.
Even if one ignores the classical economists’ institutional thoughts and begins with Veblen, Davidson’s description is inaccurate. First, it ignores “New Institutionalism” (analogous to ignoring Post-Keynesian analysis by economists like N. Gregory Mankiw). Oliver Williamson was awarded the Nobel Prize in Economics in 2009 “for his analysis of economic governance, especially the boundaries of the firm.” Williamson is a very conservative leader of the New Institutionalists and the award was for his theories of institutions.
Second, John Kenneth Galbraith was not the “last prominent” institutionalist. He played a significant role in economic thought for many decades after the 1950s.
Third, one of his sons, Jamie Galbraith, is a prominent institutionalist scholar. UMKC has multiple experts in institutional thought.
Fourth, large swaths of “government economic policy” are shaped by institutional thought. Many of this believe that this has proven disastrous because the proponents of institutional thought pushed policies that eviscerated vital institutions that had served this nation brilliantly and replaced them with ideological constructs that proved severely criminogenic. Prominent examples of these policies include the three “de’s” – deregulation, desupervision, and de facto decriminalization under the New Institutionalist claim that “markets” “self-regulate.” The changes in executive and professional compensation have also proven criminogenic. Indeed, they create the ability for the CEO to generate a “Gresham’s” dynamic and generate endemic fraud. The adoption and championing of subordinated debt as capital was a product of (failed) New Institutional policy recommendations. The repeal of Glass-Steagall, the adoption of the Commodity Futures Modernization Act (which created the infamous black hole for many financial derivatives), the insane gutting of bank capital requirements in the Basel II accord, and the “Reinventing Government” mandate by the Clinton-Gore administration that financial regulators treat the banks and bankers as their “customers” are all (terrible) products of policy recommendations prompted by New Institutional theories.
Conversely, the reregulation and the resupervision of the savings and loan industry – and the creation of an effective criminal justice response to the elite frauds driving the second (and vastly more expensive) phase of the S&L debacle was the product of institutional economic policies. I led that effort at the staff level and drew on my training in institutional economics at the University of Michigan. Those actions took place in the 1980s and early 1990s. Similarly, Elizabeth Warren’s promotion of the Consumer Finance Protection Agency was the product of her sophisticated understanding of institutional design.
Behavioral economics is frequently combined with institutional economics. Policy debates about whether and how to create systems to “nudge” customers typically represent a fusion of both fields of study.
Davidson gets something important about macroeconomics partially correct, makes a false comparison, and then gets something far more critical about it profoundly incorrect.
[T]the so-called mainstream of economic thought, based on Keynesian models, is supported by decades of data, peer review and argument. Many noneconomists think of the fundamental economic divide between Democrats and Republicans as being between government-loving Keynesian Democrats on the one hand and, on the other, Republicans who extol the free-market principles….
Over the last 75 years, the actual people designing Republican economic policies — figures like Martin Feldstein, Gregory Mankiw, Glenn Hubbard — have nearly all relied on Keynesian principles and models of the economy.
Well, no. First, the implied comparison is that “the … mainstream of economic thought [is] based on Keynesian models.” Again, this ignores the fact that there is, and can be, no “Keynesian model” of microeconomics. No one claims that microeconomics, which includes institutional analysis, lacks “decades of data, peer review and argument.” Davidson’s implied comparison of macro with microeconomics is false.
Second, yes, there are “decades of data, peer review and argument” by mainstream economists. The obvious next question is whether the mainstream economic models and the micro theories accurately predict economic results. Recall that Davidson wrote his article on April 13, 2016. To put it gently, Davison has every reason to know that mainstream economic theories of micro and macro have failed repeatedly and to a degree that would be a scandal in any real science. Instead, his column seeks to convey the opposite impression that mainstream economics has reached a triumphal fusion supported by mountains of data and an exemplary peer review process.
The macroeconomic mainstream in academia demonstrated that its models are incapable of predicting or even recognizing the largest bubble in world history. Narayana Kocherlakota, a famous “freshwater” macro academic became President of the Minneapolis Fed, the most extreme regional bank adhering to the freshwater macroeconomics school, and described in 2010 the extent of the mainstream’s failure.
I believe that during the last financial crisis, macroeconomists (and I include myself among them) failed the country, and indeed the world. In September 2008, central bankers were in desperate need of a playbook that offered a systematic plan of attack to deal with fast-evolving circumstances. Macroeconomics should have been able to provide that playbook. It could not. Of course, from a longer view, macroeconomists let policymakers down much earlier, because they did not provide policymakers with rules to avoid the circumstances that led to the global financial meltdown.
Third, the reasons the mainstream macro academics gives for the repeated, abject failure of their models is their failure to model accurately institutions. Dr. Athreya, the Richmond Fed’s Research Director, states that the models fail because of two central errors.
“[A]symmetric information and limited commitment” “are economists two ‘usual suspects’ in creating problems for decentralized trade.”
By “decentralized trade” he means markets. Athreya asserts that the interaction of these twin scourges is minimal in “spot” markets, but can prove immensely destructive in “IOU” markets. The secondary mortgage market transactions and sales of collateralized debt obligations (CDOs) demonstrates that even in the supposedly most ideal “spot” markets “control fraud” can become epidemic.
By “limited commitment” he means the inability of a party to a transaction to give credible, effectively enforceable commitments that will protect the counterparty from loss in the event that the first party breaches. That inability, of course, leads to increased defaults and greater losses upon default and increases the incentive to defraud and abuse counterparties. The ultimate form of abuse of “asymmetric information” and “limited commitment” is “control fraud,” which is one of the critical institutions in explaining hyper-inflated bubbles and financial crises. It turns out that mainstream economists now admit that their failure to understand the perverse incentives their institutional policies caused.
Similarly, Kocherlakota cited three underlying conceptual difficulties as producing the failure of mainstream macro models.
I highlight three particular weaknesses of current macro models. First, few, if any, models treat financial, pricing, and labor market frictions jointly. Second, even in macro models that contain financial market frictions, the treatment of banks and other financial institutions is quite crude. Finally, and most troubling, macro models are driven by patently unrealistic shocks.
Each of these weaknesses arises from a failure to understand institutions. “Frictions” are mainstream macro euphemisms for models that are predictive failures because they are poorly specified. A common example they use is that wages are “sticky,” and the explanation for that is institutional factors. Kocherlakota even uses the word “institutions” to describe the failure of the models to specify correctly how financial institutions operate. Third, the “patently unrealistic [technological] shocks” that mainstream macro uses as fudges to hide their models’ routine predictive failures depend on institutions (e.g., “network analysis” is all the rage in mainstream analyses of technology). The actual shocks that mainstream macro does not use in its models are also overwhelmingly institutional, particularly epidemics of control fraud.
Fourth, there are additional fatal model failures that mainstream macro does not admit. Mainstream macro’s dynamic stochastic general equilibrium (DSGE) models are based on the premise that mainstream micro “general equilibrium” (Arrow, Debreu, and McKenzie (ADM) models) are correct. ADM models are nonsense, so mainstream macro DSGE models are nonsense. Athreya admits that mainstream economists implicitly assume an “ADM god” (Athreya 2013: 103) (I am not making this up) because otherwise the standard economic assumptions would not produce equilibrium and would not maximize wealth or utility.
Those standard economic assumptions are explained by Athreya. Note how they implicitly predict massive control fraud unless governmental regulatory and law enforcement institutions are vigorous in constraining the CEOs’ perverse incentives.
[M]odern macroeconomics is almost, but not perfectly, dystopian. It remains maximally cynical about the behavior of people and the corporations they run. It presumes that large firms are aware that they are large, will monopolize if they are allowed to, and will exploit policymakers’ inability to commit to not bailing them out in bad eventualities. It implies that people will employ any means they know of to avoid taxes. And so on.
[Therefore], there can be no presumption whatsoever about outcomes being efficient (Athreya 2013: 357).
Note that Athreya recognizes that the key is control fraud –by focusing on “the behavior of people and the corporations they run.” He also admits that mainstream economic theory “implies that people will employ any means they know of to avoid taxes” and create illegal monopolies. It implies that CEOs will routinely use the firms they control as “weapons” to commit crimes that will enrich the CEO unless government institutions are so vigorous and effective that they counter the perverse incentives mainstream economists’ compensation policies have created for CEOs – and which CEOs in turn can employ to suborn internal and external “controls” and corrupt their employees. The reader can now see why mainstream economists are forced to postulate an “ADM god” that infallibly restrains the dystopian CEOs. Theoclassical economists, of course, do not actually believe there is an ADM god, which is why they make the assumption implicitly rather than explicitly. The alternative to assuming an ADM god is for mainstream economists to state that effective government institutions are the most essential requirement to protect us from dystopian CEOs who, under standard economic assumptions, will “employ any means” – including criminal – that increase their income. Theoclassical economists’ standard assumptions therefore, logically require that the only way to achieve a well-functioning economy is for the government to vigorously guard against (and prosecute) the crimes of the most powerful CEOs. Theoclassical economists’ intense anti-governmental ideology does not allow them to go there.
Fifth, mainstream macro proponents like Athreya (p. 340) strongly support Bernie’s policy that it is essential to get rid of the systemically dangerous banks, which he aptly describes as holding America for “ransom.” Again, Davidson is wrong to assume that the only economic policy economists have is macro policy.
Davidson is Wrong about “Trade-Offs”
Davidson’s big point that he wants readers to believe is that mainstream economists have long reached a consensus on the inherent nature of “trade-offs.” Davidson wants the reader to believe that there is a consensus based on “empirical data” and that Bernie ignores the experts and the data and assumes out of existence reality – making his plan fanciful. The subtext is that Republican and Democratic macroeconomists share the same beliefs because the “empirical data” proves their “similar models” to be correct. Davidson wants you to believe that the supposed consensus is not a matter of anti-governmental dogma on the part of economists, but rather simply devotion to truth and the scientific method.
[W]hen they make their arguments, these economists are using similar models built around the same sets of empirical data. And one common assumption of those models is that government intervention in the market tends to come at some cost. Offering free health care and college to everybody will cost money that needs to be raised in other parts of the economy. Some might embrace those trade-offs, others might reject them, but economists agree that there are trade-offs, that the realistic possibilities of our economy have constraints.
None of these statements is accurate. First, it is not accurate to say that macroeconomists (notice Davidson again acts like only macroeconomists exist) “are using similar models.” Davidson leaves his readers with the impression that the financial world uses mainstream academic’s macro (DSGE) models. Financial firms’ macroeconomists, however, view DSGE models as failed and useless. They virtually never use DSGE models. Instead, they use variants of Keynesian models in which demand for goods and services is a paramount features. DSGE models generally ignore demand and are not “similar” to the models used in finance. Again, the models used in finance have far better predictive records than DSGE models, so if Davidson were correct that academic macroeconomists all believed in using the DSGE models that have been consistently falsified by reality then that would demonstrate that mainstream academic macroeconomists were wholly unscientific.
Second, DSGE models are not “built” on the basis of “empirical data.” They are weirdly theoretical and bear no resemblance to reality. Only the fudge factors (the “patently unrealistic shocks”) are empirical – and after the fact. The DSGE models’ predictions fail regularly and substantially. The DSGE model makers then go in and make new arbitrary changes to the (arbitrarily) assumed “shocks” to try to come closer to mimicking actual results. This is a mockery of the scientific method, which is why the DSGE models have failed for decades.
Third, freshwater DSGE models specify, contrary to the “empirical data” what is essentially a balanced budget requirement for nations like the U.S. that have freely floating sovereign currencies and borrow only in their own currency. This specification creates an artificial trade-off that typically does not exist in the real world. My UMKC colleagues and the experts they work with in finance such as Warren Mosler, are among the world’s leading experts in Modern Monetary Theory (MMT). Unlike DSGE modelers, MMT scholars begin with assumptions that are empirical and based on how sovereign currencies actually function.
I will leave the heavy lifting on MMT to my colleagues. Randy Wray has just done a column in New Economic Perspectives addressing the trade-off claim. To reprise two key points, any “trade-off” exists when real resources are so limited that they are fully employed. It does not apply to what Davidson asserts – “money.” He also thinks money must be “raised” (presumably through taxation), so he does not understand how sovereign money is created.
Fourth, the divisions among macroeconomists are typically ideological. Kocherlakota makes a sly admission.
Given the primitive state of computational tools, most researchers could only solve models of this kind. But—almost coincidentally—in these models, all government interventions (including all forms of stabilization policy) are undesirable.
Yes, “almost coincidentally.”
Fifth, Davidson seems to be arguing that government action represents a net cost to the economy.
[G]overnment intervention in the market tends to come at some cost. Offering free health care and college to everybody will cost money that needs to be raised in other parts of the economy.
Davidson’s examples are particularly poor exemplars of the point he seems to be trying to make. Most economists believe that government expenditures on education are one principal contributors to America’s becoming an economic superpower and helped produce very large political and social gains. Notice that Davidson assumes facts that are often incorrect when it comes to education and health. There was no “market” access for most Americans to education and only very limited access of most Americans to advanced medical care for much of our history. The government’s provision of education and health for most Americans was not an “intervention” in an existing “market” so much as it represented the creation of a public market for education and health for the vast bulk of Americans.
Institutions, information, and infrastructure (I3) represent classic examples of government programs that produce widespread growth while bad institutions created by theoclassical microeconomists’ policies (such as the three “de’s”) are leading causes of massive losses of GDP and jobs and increasing inequality. Economists have estimated the U.S. loss of GDP through the course of the Great Recession and the long recovery at $24.3 trillion. A trillion is a thousand billions. The lost GDP in the eurozone is far larger because of the disastrous austerity dogma spread by theoclassical macroeconomists that dominate eurozone policy.
Athreya, a strong supporter of freshwater macro, agrees that creating effective institutions that restore the rule of law are vital to countering the negative externalities caused by elite fraud and global climate change.
A good deal of macroeconomic work on the crisis aims to clarify why privately optimal arrangements, particularly debt, can create ex-ante and ex-post inefficiency.
Hint to theoclassical economists: maybe the private arrangements are not remotely “optimal” because there is no ADM god in the real world to prevent bankers from ripping off and defrauding their customers.
Davidson has, perhaps unintentionally, helped illustrate why the ideological constraints of orthodox micro and macroeconomists have caused them their predictions and policies to fail. Two of the worst mistakes they made were related. The scientific study of institutional economics inherently requires scholars to take reality seriously and try to develop empirical methods that capture reality. Orthodox economists’ DSGE and ADM models are so embarrassing in terms of reality and predictive ability that they can be saved only by the implicit assumption of an “ADM god” invented by theoclassical economists to assume out of existence the frauds and abuses that make their ADM and DSGE models – under their own “dystopian” assumptions – fantasies. The ADM god is the only thing that can save the theoclassical economists’ dogmas. Note also that their ADM god implicitly acts infallibly and with zero “trade-offs.”