Welcome to BIG, a newsletter about the politics of monopoly. If you’d like to sign up, you can do so here. Or just read on…
This morning on Bloomberg TV, I watched a clip of economist Larry Summers making an argument for what to do in the event of a recession. Couched in the technical terminology was something profound about how we make political decisions. So today I’m going to write about the point of economics, as a discipline.
So I Was Talking to Olivier Blanchard…
This morning, on Bloomberg TV, Larry Summers said he was musing on how to deal with recessions, and felt that the government needs to spend more money to address the next downturn. “I was talking to Olivier Blanchard…” he said, Blanchard being his economist colleague for four decades. They came up with something called ‘semi-automatic stabilizers.’
There are a many ways to talk about how to run a society, and economists have their own language. They talk about ‘monetary space’ when they mean printing money and ‘fiscal space’ when they mean the government spending money. Summers is trying to create a new word, in this case ‘semi-automatic stabilizers,’ which are ways of changing “tax or spending measures triggered by the crossing of some statistical threshold,” like shifts in output or unemployment.
Are ‘semi-automatic’ stabilizers a good way to address recessions? We already have automatic stabilizers like unemployment insurance, income/corporate taxes, and welfare. They work and we should make them stronger. I don’t have a strong view of semi-automatic stabilizers one way or the other, because I don’t know what ‘semi’ means in the context and institutional details matter. I can easily imagine them being gamed to place economists in a position of authority over fiscal policy they way they run monetary policy, but they could also be done simply and well. That said, while technocratically organized constraints on fiscal policy may or may not serve useful ends, they certainly can shift authority from elected leaders to economists. And I suspect for Summers, that’s the point.
More important than the proposal is the rationale. Summers argued in favor of ‘semi-automatic’ stabilizers not on the economic merits but because Congress’s decision-making is too slow to handle recessions. He is clearly burned from the financial crisis, when he proposed a stimulus undersized relative to the economic hole needing to be filled. As Reed Hundt pointed out in A Crisis Wasted, the economists in the White House knew they were misleading the public and Congress about the depth of the nation’s problems. So when they asked Congress for more spending, Congress refused. This wasn’t a Congressional procedure problem of speed, as Summers alleges, but a political problem. One important reason Congress refused more spending is Summers and the White House never asked for what they needed in the first place. Having lied to Congress about the problem, Congress didn’t trust the administration when they asked for a bit more. And frankly, the White House was split; White House economist Peter Orszag and Obama himself sought to reduce the deficit.
Summers’s point in proposing ‘semi-automatic stabilizers’ is to blame Congress for a problem Summers helped cause. Summers is open that he’s not trying to solve an economic problem, he’s just trying to get around Congress, which he thinks shouldn’t be making key decisions about resources in the economy. One thing that is clear is we definitely should have had a better response from Larry Summers during the recession, when he ran economics in the White House and screwed it up. And if there’s a choice between having economists choose how to spend money or democratically elected leaders, it should probably be the latter.
And this brings me to the point of economics, which has taken me a long time to understand. There are many economists who focus on trying to uncover important truths about the world, and there are many economists who seek to serve concentrated capital. There are smart ones, and dumb ones. But truth or falsehood, or empirical rigor, is besides the point. The point of economics as a discipline is to create a language and methodology for governing that hides political assumptions from the public. Truly successful economists, like Summers, spend their time winning bureaucratic turf wars and placing checks on elected officials.
Chameleons in Economics
Let’s start with a basic question. Is it the job of economists to understand the world accurately? The answer is far from clear. As finance professor Paul Pfleiderer notes, many economists use models that are chameleons, designed to launder political assumptions about the world through such aesthetics as “mathematical elegance, subtlety, [and] references to assumptions being “standard in the literature.”
More broadly, prominent leading economists just get things wrong, big things, with no impact on their standing in the profession. This indifference to empirical results is mirrored in the indifference offered by economists to the claims and arguments of non-economists outside the profession, whose views are simply relevant.
I’ll bring you through a few examples.
(1) In 2004, Ben Bernanke gave a speech called the Great Moderation, in which he lauded the placid prosperity economists had brought forth since the 1980s. “One of the most striking features of the economic landscape over the past twenty years or so,” he argued, “has been a substantial decline in macroeconomic volatility.” As Bernanke was speaking, the FBI was warning of an “epidemic” of mortgage fraud. Needless to say a few years later there was the most gruesome financial crisis in 80 years.
(2) In 2017, antitrust economist and Google consultant Carl Shapiro described his tenure as the head economist at the DOJ antitrust division, and explained why the division had brought no monopolization claims in the period he was there.
First, I can say from personal experience that when I was the chief economist at the DOJ during 2009-2011, the Antitrust Division was genuinely interested in developing meritorious Section 2 cases, and we were prepared to devote the resources necessary to investigate complaints and other leads, but we found precious few cases that warranted an enforcement action based on the facts and the case law.
Is that right, that there were literally no cases of monopolization in the American economy from 2009-2011? It’s hard to imagine so. And yet, this is Shapiro’s claim.
(3) In the middle of the 2000s, millions of Americans were complaining that they were losing their jobs, which were being offshored to China. Economists largely rejected these complaints. For instance, in 2004, Bush official and economist Greg Mankiw argued offshoring was largely a good thing and anyone who said otherwise was a quack. In asserting this he was simply reflecting the economic consensus. Ten years later, several economists pulled apart this consensus coining the term “China shock” in 2016. In other words, it took ten years for those within the discipline of economics to even be able to hear the views of millions of normal Americans being injured, and their views had to be laundered through other economists.
(4) Today, many Obama era antitrust officials are discussing the need for more assertive antitrust enforcement, arguing they were hamstrung by limited legal tools while in power. That is not what they said at the time (much as Larry Summers today argues for more spending but in office sought different policy choices). In 2016, the consensus of antitrust economists and lawyers - as illustrated by the American Bar Association Antitrust Transition Report - was that competition law was quite robust.
Antitrust figured prominently in the 2016 Election: for the first time in recent memory, both major parties prominently featured their respective viewpoints on competition and consumer protection policy. Campaign commentary included sharp criticism of an alleged absence of vigor and overall ineffectiveness in current patterns of antitrust enforcement, with comments calling for sharp, even radical reorientation of enforcement policy, especially in the review of proposed mergers and treatment of large industrial firms. As will be seen in this Report, the Section does not share these views about the current state of antitrust enforcement policy. To the contrary, … the Section’s view is that the Nation’s system of competition enforcement has been in good hands, [and] that an arc of continuous improvement and advancement can be discerned that stretches back over many years and multiple administrations.
So there we go. If the goal of economics were to ascertain truthful views about the world, if economics were as its proponents offer, a ‘science,’ then one would remark on the lack of self-policing within the profession. Of course, given that there is limited self-policing at best and the top practitioners in the field are routinely wrong about fundamental questions, we can conclude that uncovering truth may be an incidental outcome of the practice of economics, but it is certainly not the goal of the discipline.
Methodological Biases in Economics
There are three main problems with economics as a ‘science’ that can guide public policy choices. The first is that it is a post-mortem discipline. Economists often assert we need data before drawing conclusions. Economist Thomas Phillipon noted this in his book on the institutional basis of markets that an economist was like that of Sherlock Holmes, asserting ‘data data data, I cannot make bricks without clay.” And yet, there was no data in 2000 when the U.S. changes its policies vis-a-vis China, because the consequences were in the future. There’s nothing wrong with being a study of the past that has a specific quantitative framework, as long as there is a genuine acknowledgment that there’s no science here and projections have no scientific validity whatsoever.
The second is that using economics to make judgments about the world can be extraordinarily costly and exclusionary. This may or may not be a big deal when considering macro-economic forecasting, but when economics becomes a key part of institutional legal arguments it shades who can use the law to protect their rights. For example, showing that someone robbed me by breaking into my house requires evidence and common sense. But bringing an antitrust lawsuit showing someone robbed me by excluding me from a market often requires millions in economics consulting services. If you don’t have that money, the law becomes meaningless.
The third is that an obsession with quantifying leads to political control by those who have access to data. A well-known example is famous economist Alan Krueger, who was paid by Uber and then wrote widely circulated scholarship based on internal Uber data about the corporation’s wage setting terms. But it’s broader than just one company, most of the big tech platforms work with economists, giving these powerful corporate entities a measure of political control over lines of research. Beyond tech, it’s actually quite hard to get information on a whole host of practices in the economy. For example, the Trump administration had to battle hospitals just to get them to disclose their official price list for different procedures (which isn’t even real considering the extent of secret discounts and rebates throughout the industry).
How Does Economics Work?
The actual goal of economics as a discipline is to embed itself as a governing language in our institutions of power. There are four institutions in which this takes place domestically.
(1) The Congressional Budget Office: This obscure bureaucratic agency does modeling for all legislation in Congress, projecting economic impacts and observing whether legislation will increase or lower the deficit. Importantly, the House and Senate have delegated to CBO an important chokepoint in the legislative process; bills that raise or lower deficits as per CBO projections be be held to points of order, which is to say, members of Congress have to affirmatively vote to ignore what is portrayed as the scientific truth.
CBO seems to get things wrong in ways that privileged concentrated capital and a certain form of austerity-driven politics. Here are two simple examples. First, CBO for most of the post-2009 era assumed, based on opaque and reactionary economic modeling, that interest rates would soon snap back to 5%, which effectively meant that spending more money through tax cuts or spending increases, as many legislators wanted to do to help their constituents would be quite costly. Turns out interest rates didn’t come back to 5%, and the assumptions behind those interest rate models had hidden political biases favorable to concentrated capital.
In other words, CBO prolonged the recession through bad projections. CBO’s response to criticism is that they basically grabbed interest rate projections from mainstream economists at prestigious universities and investment banks. The top economists got it wrong, so CBO got it wrong. It’s very much ‘nobody got fired for buying IBM’ except in this case what is happening is that legislators are hearing that the ‘scientists’ are falsely telling them what we can and can’t afford based on models that hide their politics in complex math.
Another example, perhaps a simpler one, is that CBO scored the effect of derivative deregulation legislation - such as a bill known as the Swaps Pushout Provision in 2014 - at essentially zero. Such legislation may cause a financial crash, and that would cost the government money, but since CBO can’t project how much or when, and since economists don’t have models for it, they say it costs zero. In other words, spending money through the regular budget gets subject to points of order, but spending money by shifting risk onto the public balance sheet by letting banks gamble with our money doesn’t. Guess which one Congress regularly enables?
(2) The Office of Information and Regulatory Affairs - Last month, an anonymous government lawyer wrote an issue of BIG on how economists at this little known agency probably weakened a law against prison rape by subjecting the rule to a process called ‘cost/benefit analysis.’ You can read how that worked, but the gist is that such cost/benefit analysis is another place where hidden political choices are inserted into policy through opaque economic models. OIRA is part of the White House Office of Management and Budget, and they review nearly every regulation put out throughout the government. Cost/benefit sounds good, but it’s another choke point where democratically elected policymakers delegate power to economists.
(3) Federal Trade Commission/DOJ Antitrust Division - I noted above the problem with economics and antitrust. To bring a case, plaintiffs have to spend a few million dollars just for an economist to model the harm. The joke is “I know it happens in reality, but show me in theory.” The demand for quantitative models - which are often garbage and yet trusted by judges trained to believe in economics expertise - is just a smokescreen for replacing the rule of law with the rule of economists. The net effect is that Facebook, Google, and Amazon bought hundreds of companies over the past twenty years through three administrations, and there hasn’t been a single merger challenge. Or to be more granular, John Kwoka did a merger retrospective analysis, and observed that the FTC has systemically allowed too many mergers in concentrated industries, which induced significant price hikes across multiple sectors. And as we saw above, noted economist Carl Shapiro, from 2009-2011, couldn’t find a single instance of monopolization in the American economy.
There’s also a level of intimidation toward elected officials by the cult of the neutral technocrat. When Elizabeth Warren pointed out we have to break up Google, Facebook, and Monsanto, several of her opponents argued that singling out specific companies is ‘Trump-like’ and a violation fo the spirit of the rule of law. Such an argument is quite anti-democratic. The FTC Act Section 6(D) allows the President or Congress to order an investigation into a specific company, because a large corporation is a matter of public concern. And yet there’s an attempt to science-ize what are inherently political decisions of social concern.
(4) The Federal Reserve - As Paul Conti-Brown showed in his excellent book on the Fed, Paul Volcker established the social norm that politicians shouldn’t meddle with monetary policy, that printing money is technocratic. This is couched in the framework of ‘independence.’ And yet, independence is yet another mechanism to move policymaking authority from elected leaders to economists.
Economists have no credible claim to be good at central banking. I noted Bernanke’s problematic assertion in 2004 that we were in a period of a great moderation. In 2005, Raghuram Rajan at a Jackson Hole conference of central bankers noted hidden risks in the financial system, which nonprofit advocacy groups dealing with mortgage origination in bubble-y markets such as Las Vegas had recognized as early as 1999. Larry Summers mocked Rajan, praising Alan Greenspan’s stewardship of the Federal Reserve.
It hasn’t really improved. A few months ago, Fed Chair Jerome Powell, with his complex matrix of data points and his legions of economists, got into a highly publicized argument with Donald Trump, over interest rates. Trump criticized Powell for potentially tightening at a moment when the economy was slowing, saying you have look beyond data and ‘feel’ the market. Powell reversed himself after data finally came out showing Trump, with his gut feel, was right.
The point of this essay isn’t to argue economists are corrupt or bad or anything of the sort. There are many wonderful economists, and I sometimes - even in this essay - rely on their work. The point is to note that doing useful work is only incidental to the discipline of economics, whose real institutional goal seems to be the development of an exclusionary language and methodology, and then the embedding of economists into policymaking positions where they have a good chance of overriding the judgment and authority of elected officials.
There is a tension between how economists have arrogated policymaking ability within our governing institutions, and the need for democratic accountability. The way to address this dynamic is to put economists back into a position where they are offering advice rather than making policy.
Institutionally, this means making politics explicit instead of hiding political assumptions in models. To that end, it might make sense to split up CBO into a Democratic and Republican versions, and remove points of order from the legislative process. It probably means a return to direct Congressional or executive direction of monetary policy, a means of governance that existed through most of the 19th century and from 1935-1951. It means eliminating the Bureau of Economics at the FTC (as well as the newly created Office of Economics and Analytics at the FCC), and placing economists in a supportive role under lawyers, as well as developing a style of lawyering that doesn’t rely on economic expert witnesses. And finally, it means ending the rigged game of cost/benefit analysis as a chokepoint on promulgating regulations.
In 1938, Franklin Delano Roosevelt warned Americans of rising forces abroad. “History proves that dictatorships do not grow out of strong and successful governments,” he said, “but out of weak and helpless governments.”
What has ultimately happened with economists arrogating policymaking authority is that our government has become weak and unresponsive to the concerns of its citizens. Economists and non-economists must address this problem by focusing on renewing the ability of ordinary people to have their concerns heard above the din of error-ridden charts. Otherwise, autocratic forces will be happy to take advantage and make the case that democracy can’t deliver.
Thanks for reading. And if you liked this essay, you can sign up here for more issues of BIG, a newsletter on how to restore fair commerce, innovation and democracy. If you want to really understand the secret history of monopoly power, read my book, Goliath: The 100-Year War Between Monopoly Power and Democracy.
UPDATE: I edited the opening bit on Larry Summers because it could be read as making a blanket argument against rules-based programs, which was not the claim I was trying to put forward.