In his Free Market: The History of an Idea <https://www.amazon.com/dp/0465049702/> Jacob Soll fakes right, making me think I am going to read an appreciation of 20th century free-market thought:
Twentieth-century free-market thinkers… Friedrich Hayek to Milton Friedman… constituted a powerful, conservative force that foresaw the authoritarian and totalitarian dangers—on the left and the right—that lay ahead…
But no! He goes left! He brings down the hammer, in a way that would make Judah “The Hammer” son of Mattathias proud:
And yet, along with the great moral achievements and economic insights of free market thinkers came a very particular form of paranoia, ideological obsession, and myopia…. Twentieth-century orthodox free-market economists believed that pure individual desire and agency were the catalyst for all societal and economic good. In their eyes, any system that deviated from this view became suspect. It was not so much an academic position as an article of faith…
He brings down the hammer specifically on Friedrich von Hayek:
The Road to Serfdom would become the handbook of postwar free-market and libertarian economics… less a work of economic theory than a declaration of total libertarian faith…. With hindsight… the book stands out for its total lack of engagement with the realities of the postwar growth period and its fanatical vision of the state as a force of evil…
And:
Hayek saw market freedoms in a combative light, emerging from a struggle between good and evil. One either chose economic liberalism with no government, or one would be enslaved…. Hayek chose to forget that Hitler could neither have taken nor held power without the concerted support of German capitalists, who saw fascism as an attractive answer to trade unions, communism, and even social democracy…
I do understand. I, myself, see von Hayek as 40% Dr. Jekyll and 60% Dr. Hyde—great insights and very valid concerns, somehow married to a majority of crazypants views on macro, politics, and moral philosophy.
Soll’s book is a book that I very much wish that I had had when writing my Slouching Towards Utopia <bit.ly/3pP3Krk>, for he takes the von Hayek-Polanyi oppositional dialectic that I use as a major focus of my book, and projects it back past the Enlightenment and the Mediæval Era to the Classical Age of Cicero. It taught me a lot, and brought much that I had only dimly realized into focus.
Soll is 100% right in finding and stressing a great divide between what “free market” thought meant before and what it meant after 1900. There was a very large shift away from an earlier perspective traced by Soll, one in which the free market can work very well indeed when exchange is between individuals who are roughly on the same level with respect to their social power. In that view, “the free market” is only one arrow in a large quiver of alternative instruments of statecraft and human cooperation:
By contrast, twentieth-century free-marketeers saw the untrammeled market as both necessary and sufficient for… something. What was not clear. It did not seem to be any form of general human flourishing. Instead, it was, as I quoted above, in Soll’s judgment:
A very particular form of paranoia, ideological obsession, and myopia…
One way to think about this is that pre-1900 Free Marketeers wanted to free entry into the market—that people should not be bound to work for another who was their master in the sense of dominus, that people should not be prohibited from producing or selling on account of their social status, that the king should not sell monopolies. This then bled over into a concern that the king should not disproportionately tax commercial wealth in order to reward his friends, and that individual, small group, or collective entities should not themselves be allowed to constitute themselves as monopolies either.
This freedom to enter and participate in a market that was free to work was part of a general Enlightenment vision which, Soll writes, was a
vision of progress through benevolent moral discipline, education, radical science, and a worship of agriculture…
With Adam Smith being perhaps its foremost exponent. But this is not a “free market” in the sense of post1900 Free Marketeers. Indeed, the market of Adam Smith and his predecessors is a market that has to be carefully and successfully managed—not least in that the people who meet each other in the marketplace need to be on the same level, have comparable levels of social power. If not, then, as Karl Marx wrote, the market appears to be:
a very Eden of the innate rights of man. There alone rule Freedom, Equality, Property and Bentham. Freedom, because both buyer and seller of a commodity… are constrained only by their own free will… contract as free agents… give legal expression to their common will… [to] exchange equivalent for equivalent… [as] each disposes only of what is his own… in accordance with the pre-established harmony of things… for the common weal and in the interest of all.
But in reality, Marx writes, in the market system of rich bourgeois and poor proletarian:
we… perceive a change in the physiognomy of our dramatis personae…. capitalist;… [and] labourer. The one with an air of importance, smirking, intent on business; the other, timid and holding back, like one who is bringing his own skin to market and has nothing to expect but — a tanning…
It was Alfred Marshall who soft-pedaled the income-distribution and social-power worries, and also pushed forward the idea that the market did not need government regulation, for it would and could regulate itself.
Alfred Marshall’s star pupil John Maynard Keynes disagreed. Keynes saw mammoth defects in the ends the market pursued as a result of unequal income distribution and the ability of the market to regulate itself and employ resources properly, with the principal resource misuse being the scandal of depressions and the result unemployment. Keynes, however, thought—hoped?—that these defects could be repaired easily, with the lightest-hand management of the economy possible. He saw that:
the outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes…
A market economy managed by his technocratic disciples would, first, attain and maintain full employment. Then even someone who owned only his own skin would have at least some social power when he brought his self to the market—for without the worker’s hands, eyes, and brains, expensive machinery would lie idle, rusting away. Yes, the worker had to find a job or starve. But in a constant full-employment economy the boss would have to find a worker or go bankrupt.
But, second, there would be more. The full-employment monetary policy that Keynes’s technocratic disciples would manage the economy to would be a very low-interest rate economy. That meant, Keynes thought, the euthanasia of the rentier: plutocrats would only be able to use their social power as property-owners to control affairs if they spent down their capital, and then they would cease to be plutocrats.
Keynes was thus a supporter of “free markets” in the older sense. As Soll notes:
in the 1920s, he warned of a battle between communism and individualistic laissez-faire, which laissez-faire had to win. But Keynes felt that there were holes in free market theory…
Chiefly the two I noted above. Thus the state would have to take “an even greater responsibility for directly organizing investment.”
In some ways, Milton Friedman bought Keynes’s argument about the inability of the economy to regulate itself. But Friedman thought that the central bank could do all the regulation and management needed. By defining whatever monetary policy achieved constant full employment as a “neutral” monetary policy that was be definition the opposite of government intervention, Friedman hoped to win a game of intellectual free-card-monte, in which what Keynes called government intervention, regulation, and management was redefined as the real hands-off.
This, I think, led to a lot of confusion.
Soll concludes his book:
Free individual action is essential to the dynamism of the market, but it alone does not guarantee the economy’s steady functioning…. We would do well… to return to… Cicero… [for] lessons… Wealth was only good, Cicero thought, insofar as it could be used to support constitutional government, civil peace, and decorum. More important to him than riches were the principles of living in harmony with nature, cultivating learning and friendship, and doing the hard work of ethical stewardship. Faith in the market alone will not save us, but hewing to these old virtues just might…
In short, be pragmatic: The market should be free to the extent and in the direction that such freedom leads to widely distributed prosperity, and wealth should be pursued not as an end in itself but a means to the political and societal good.
Should any of us disagree?
Brad DeLong & Ezra Klein: How the 1970s transformed American politics: <https://www.nytimes.com/2022/11/04/opinion/ezra-klein-podcast-bradford-delong.html>
With a knick-knack, patty-whack, give the dog a phone…
This is a judgment on one of the intellectual communities I have been contributing to—or, rather, attempting largely unsuccessfully to contribute to—throughout my career. It is funny, in a brutal way. Or, perhaps, it is brutal, in a funny way:
Daniel J. McDonald & Cosma Rohilla Shalizi: Empirical Macroeconomics & DSGE Modeling in Statistical Perspective: ‘Daniel graduated…. Smets and Wouters (2007)… needed both a lot of programming time and a lot of computing time to churn through thousands of variable swaps and tens of thousands of fits to simulations. We both got busy with other things…. But what we can tell you now, with great assurance, is that:
Even if the Smets-Wouters model was completely correct about the structure of the economy, and it was given access to centuries of stationary data, it would predict very badly, and many “deep” parameters would remain very poorly estimated;
Swapping the series around randomly improves the fit a lot of the time, even when the results are substantive nonsense.
The bad news is that even if this model was right, we couldn’t hope to actually estimate it; the good news is that the model can’t be right, because it fits better when we tell it that consumption is really wages, inflation is really consumption, and output is really inflation…
Recognize that this DSGE model language has for a generation been how highbrow domestic macroeconomists (attempt to) communicate with one another.
One way to look at it is that DSGE models are members of the class of macroeconomic models of the domestic economy that are, as Chris Sims taught me now long ago, are sufficiently flexible that their structure places no restrictions on their forecasts at all. Their forecasts are thus those of VAR—vector autoregression—models, and fit as well or as poorly as VAR models fit. Their estimated structural parameters are whatever values are needed to, when processed into the reduced form, generate the best-fitting coefficients of the VAR. Beyond that role as proto-VAR coefficients, they are thus as close to being pure noise as can be found in this Fallen Sublunary Sphere.
Chris Sims originally made this as a critique of the “structural” models of the 1970s. But it applies to DSGE models as well.
There was one paragraph in McDonald and Shalizi that made me especially wince. It was at the end of their “replicating Smets-Wouters” section:
Table 1 presents the posterior mode [maximum likelihood estimate of the Smets-Wouters model parameters “from our simulated annealing method, which stochastically explores the likelihood surface in a principled manner”]. Note first that some of the parameter estimates are similar to those presented in Smets and Wouters (2007)… while others differ dramatically. However, comparing the likelihood of of our estimated parameters to those in Smets and Wouters (2007), our fit is significantly better. For our dataset, the penalized negative log likelihood of the parameters is 1145 compared to 1232…
In short, the McDonald-Shalizi computer found a parameter vector that the Smets-Wouters model thinks is 60,000,000,000,000,000,000,000,000,000,000,000,000 times as likely as the parameter vector found by the Smets-Wouters computer back in 2007, and that Smets-Wouters then reported as the “maximum likelihood” vector of the parameters.
“Our fit is significantly better”, indeed.
And then there is the feeding-the-model-data-generated-by-the-model-to-see-if-it-can-recovery-the-truth test:
Variability declines as the size of the training set increases, though not the average…. It improves markedly as the training set increases to about 400 observations (=100 years) but then plateaus…. As we get more and more data, we can not predict new data any better. This indicates one of three possibilities: (1) that with about 400 observations, we can estimate the parameters nearly perfectly, (2) that the model is poorly identified—some parameters will simply never be well estimated, but we can predict well anyway, or (3) the data are so highly correlated that the range of training observations we consider is far too small—we actually need millions of observations in order to see a meaningful decline in out-of-sample predictive performance….. The blue line in Figure 2 is the out-of-sample mean prediction error for the true parameters. The test error is not getting any closer… plateauing slightly above the baseline by about 400 training points. This seems to suggest that explanation (2) is accurate: even with more data, we will never be able to recover the true parameters, though we get some improvement in predictions relatively quickly…
Yes, we today have roughly 1,000 times more computer power at our disposal than we did in 2005. Yes, high-dimensional likelihood functions are positively Lovecraftian. But at least Claudius Ptolemy could fit his model and use it to accurately predict. And his structural parameters—the commonality of the epicycle vectors of the outer planets with the deferents of the inner planets and the sun, plus the sun’s lack of an epicycle—gave powerful clues to a better model.
Jamelle Bouie: The Attack on Paul Pelosi Has Unmasked the G.O.P.: ‘An important part of our politics has been the pretense that our leaders care about appearances, even as they fight to gain and hold power by any means necessary. Abraham Lincoln was both a bare-knuckled partisan brawler and a sagacious, broad-spirited political leader. So were many of our most revered and respected presidents, from Thomas Jefferson to Franklin Roosevelt and beyond. From the beginning, Americans saw virtue — whether real or feigned, sincere or performed — as a key ingredient in the practice of republican self-government…
Jordan Schneider & Irene Zhang”: US-China Chip War with the Chip Avengers: ‘What will the Biden administration’s new export controls mean for the US and Chinese semiconductor industries as well as the future of the US-China relationship?… I assembled the Chips Avengers: Reva Goujon (Rhodium Group), Jay Goldberg (Digits to Dollars), Doug O’Laughlin (Fabricated Knowledge), and Martin Chorzempa (PIIE)…. China can produce 14-16nm logic chips. Biren can produce GPUs at these thresholds. YMTC can produce NAND at this threshold too. What’s very interesting is that they did not set the thresholds at an aspirational level for China. They set them at levels that China already is able to do, and it’s cutting off their ability to do it. The fact that China’s already at this level tells you that the Biden administration was in a bit of a rush. They felt like they had to do this now.
Patrick Collison: CEO Patrick Collison’s email to Stripe employees: ‘You might reasonably wonder whether Stripe’s leadership made some errors of judgment. We’d go further than that. In our view, we made two very consequential mistakes, and we want to highlight them here since they’re important: (1) We were much too optimistic about the internet economy’s near-term growth in 2022 and 2023 and underestimated both the likelihood and impact of a broader slowdown. (2) We grew operating costs too quickly. Buoyed by the success we’re seeing in some of our new product areas, we allowed coordination costs to grow and operational inefficiencies to seep in. We are going to correct these mistakes…
