In Defense Of Academic Economics

Economists are definitely moronic on average on a number of issues, most notably immigration and human biodiversity, and prioritizing economic growth as the only measure of human welfare.

But all these objections can equally be expressed in the language of economics. The first says that human capital is partially heritable and unevenly distributed, and there are positive externalities from individual human capital and from cultural or social cohesion. The latter says that there are more components to the utility function than just consumption, and that social interactions may enter a utility function directly.

And in doing so, phrasing the critique in this language implicitly preserves something important. If the claims of some economists are stupid, they are also bad economics. Economics aims to describe the world around us, and if it fails on those terms, it deserves to be criticized on those terms. The aim, however, is defensible. Even if one believes that there is a progressive influence of poor reasoning, economics is a field worth defending, rather than ceding wholesale to the enemy.

But I don’t think this battle in primarily political terms is actually what’s going on. Many reactionaries of various stripes, if we are honest with ourselves, will admit to having a somewhat contrarian disposition that takes some pleasure in being right in the face of misconceived common opinion. In addition, modernity affects so many aspects of discourse that there are fields within higher education that are so corrupted as to be of questionable overall value. Anything ending in “studies” is a pretty good candidate, for instance.

But I think it is a mistake to reject economics en masse, or even to think that economics need to be wholly reinvented from the ground up. A lot of economics, even mainstream economics in its current form, would be entirely useful and usable to any reactionary statesman.

Part of the mistake comes from underestimating the breadth of opinions within economics.

Another part comes from the tendency, of which I am guilty, to only read critiques of opinions, rather than the original sources themselves, which are often a lot more nuanced than they’re given credit for. Some of these are impenetrable to lay readers. But plenty are actually quite easy to follow if you give them a try. Or even if the original is hard, you can usually find good layman’s summaries of most important concepts.

Another form of mistake comes from the instinctive tendency of reactionaries to prefer older authorities over newer authorities. The desire to read and trust old books, may serve one well in history, politics, literature and culture. But it would serve one very poorly in physics or genetics. Economics, again, is probably somewhere in the middle.

And even if some parts have been corrupted by modern fascinations with ideas like open borders, there is much in economics that has been genuinely new advancement of knowledge. The field of academic finance, for instance, essentially dates back to the 1950s. If you go looking for financial wisdom only in old books, you are likely to come up sorely disappointed.

There is a fair amount of interesting material, for instance, in Graham and Dodd’s famous 1934 textbook, a favorite of Warren Buffett’s. But anyone who tells you that this is all you need to know about investing is either a fool or a fraud. Some sample questions to people making such claims: how should I optimally combine these securities into a portfolio? Crickets. (answer: Markowitz (1952) portfolio theory) How should I value an option? Derp. (answer: Black and Scholes (1973) famous option formula). How exactly do investors form judgments and decisions? (see Daniel Kahneman’s body of work, or Richard Thaler’s). The list goes on.

And not only that, reading only old sources often misses the way that the field has actually changed over time to incorporate a lot of the early criticism.

One example of this is Austrian economics. I confess to not having read as much of this as I should. But one area where I do broadly know the Austrian critique is in the question of banking crises. In part, I take my version of the Austrian critique from Mencius Moldbug’s excellent articles, such as here, here, and here.

This particular question is worth some focus, because I quite often hear reactionaries citing this in various forms, such as the problems of fractional reserve banking and maturity transformation.

The Austrian analysis of the problems of banking is excellent. To give a very brief summary, at least of the Moldbug version, the problem fundamentally is that banks engage in maturity transformation. They take what’s known as “demand deposits” – liabilities with a maturity of zero. In other words, when you give your money to the bank, you have the option to withdraw it again instantaneously. As long as you don’t, you’re effectively rolling over your loan on a continuous basis.

Meanwhile, the bank lends out most of the funds it thus receives at long maturities, to businesses and homeowners who only promise to pay back the loans at a longer period. Hence the maturity transformation – they take liabilities with a maturity of zero and match them with assets with a maturity of several years. In addition, they keep some fraction of their deposits as reserves (hence “fractional reserve banking”). The idea is that only a small fraction of depositors will actually demand their money at any given time, as long as needs for cash are roughly independent. Hence, the bank can profitably lend out the rest, the borrowers get funds for their long-term projects, the lenders get instantaneous demand deposits with higher interest rates than otherwise, and everybody wins.

Well, most of the time. The problem arises because this equilibrium is fragile. At any one time, there aren’t nearly enough funds to pay depositors if they all demanded their funds at once, since most of them have been lent out. These loans are profitable in the long term, but not if they have to be liquidated immediately, which means that if too many people turn up at once and demand their money, the bank won’t have enough. Since it presumably pays demands in a sequential manner, if there’s an excess of demand for withdrawals, the first guys get paid off in full, and the latter guys get some smaller fraction of their claims, or maybe none at all. As a result, if you think there are going to be an excess of withdrawals, your incentive is to rush to get there first to withdraw your own money. This is called a bank run. And it means that the scheme is fragile, as any belief in the existence of a run, however arising, is enough to create the run itself. In this way, bank runs can be a self-fulfilling prophecy. So the banks periodically collapse, financial crises ensure, along with related economic problems.

So far, so good. What’s the Moldbug and Austrian solution? Moldbug quotes Mises:

For the activity of the banks as negotiators of credit the golden rule holds, that an organic connection must be created between the credit transactions and the debit transactions. The credit that the bank grants must correspond quantitatively and qualitatively to the credit that it takes up. More exactly expressed, “The date on which the bank’s obligations fall due must not precede the date on which its corresponding claims can be realized.” Only thus can the danger of insolvency be avoided.

In other words, ban maturity transformation.

If you want to lend out your assets for a year, your deposits must not be redeemable for a year. Problem solved.

The Austrians are correct in several important respects.

Firstly, they have identified a genuine economic problem.

Secondly, they have described how it operates.

Thirdly, they have proposed a solution which indeed solves the problem.

So far, so good.

Before we get into the question of whether we should follow the Austrians’ advice, I want to pose a more basic question.

Is this a critique of mainstream economics?

At the time when Mises was writing, almost certainly. But today? That’s a much harder case to make.

When I first read Moldbug’s description of the Austrian analysis of banking crises, I thought to myself, “Wait, isn’t he just describing the Diamond and Dybvig (1983) model of bank runs?”

And indeed, almost right up the end, he is.

Don’t believe me? Here’s the original paper. Or if you prefer a layman’s summary, the Wikipedia version is quite easy to follow. Have a read at least of the Wikipedia article (it won’t take long). Tell if you think that Diamond and Dybvig don’t substantially agree with Mises as to how bank runs occur.

And Doug Diamond is not some outsider in the economics profession. He’s a chaired professor at the University of Chicago, and on Clarivate Analytics’ list of potential future Nobel Prize winners. Indeed, if they give one out for banking, he’d be extremely likely to be included.

If you were snarky, you might say that Diamond and Dybvig just co-opted (or appropriated) Mises. Perhaps. Though the equally snarky rejoinder would be that responding to criticism by taking on board its critics’ best points is exactly how you’d hope a healthy intellectual field would respond.

But Diamond and Dybvig also analyze the problem with the tools of game theory, and deliver some genuinely new insights. Indeed, a major focus of their paper is on how deposit insurance can also solve the problem. If the government (which has the power of taxation) guarantees bank deposits, then depositors have no incentive to run on the bank.

Deposit insurance is actually a very important concept, even if it’s not widely studied by reactionaries. It’s a class of public good quite different from the classic lighthouse example. Because in the simple version of the theory, it doesn’t actually cost very much to implement relative to its benefits. (Practice, of course, is more complicated). The reason is that in equilibrium, the expected amount of payoffs that the government has to make to depositors due to bank runs isn’t very high, because the guarantee itself considerably reduces the chance of there being a bank run in the first place. If you know you’ll get paid regardless, there’s no longer any rush to be the first to the bank, so the fragility causing bank runs largely disappears. And governments can provide this guarantee in a way that private providers can’t, because the government has a firm ability to pay through its powers of taxation.

In terms of a type of public good, it’s as if the government agreed to build a lighthouse, and simply by announcing that, the lighthouse didn’t actually need to be provided.

And this is something many Austrians don’t seem to do very much, namely debate “ban maturity transformation” and “implement deposit insurance” as alternative ways to solve bank runs, each with their own costs and benefits.

Of course, the Austrians are right that deposit insurance comes with its own problems. The biggest of these is moral hazard and the associated regulatory apparatus needed to control this. When you guarantee deposits, banks have enormous incentives to just jack up risk, because they don’t pay the cost if things fail. So deposit insurance means that you now need an enormous monitoring apparatus to ensure that banks are properly capitalized and not taking hidden risk in their portfolios. This is a real cost, and not something to be sniffed at. The Austrians are right in thinking that this is undesirable, and probably quite costly.

The Austrians are also right in claiming that mainstream economists don’t consider enough the idea of simply banning maturity transformation. Which is true. I wish they would debate it more, since it is a solution worth serious consideration, and many fans of Diamond and Dybvig don’t seem to really ponder that this alternative solution exists.

But the Austrians also never seem to apply the same cogent critique of the cost of regulation to their own proposals. Take the Mises solution of banning maturity transformation.

It just rolls off the tongue, doesn’t it? Ban maturity transformation, and the problem disappears.

To which I say—do you have any idea how large a regulatory apparatus you would need to ensure that this policy actually takes place?

The simplest claim is to ban demand deposits. Until you realize that demand deposits are just a limiting case of a very short maturity loan that automatically rolls over unless you choose not to renew. If we ban “demand deposits”, banks will just implement “hourly loans”, or “daily loans”, or whatever term you’re not willing to ban. The problem, in other words, is the reliance on the rolling over of assets, which creates the maturity mismatch.

Even if you ban demand deposits and we end up with some minimum term length of deposits instead, how easy is it going to be to inspect the asset maturities of every asset of every bank and check that they match exactly what has been promised to depositors? Are you sure you can do this with fewer staff and resources than are currently being used to monitor bank health under deposit insurance schemes?

More importantly, the Austrians are stuck in a very early-20th-century view of how banks work. In their mind, the world can be neatly divided into banks and non-banks.

But if the problem is that mismatched assets and liabilities result in financial fragility and costly liquidations, then this problem is much, much broader.

For instance, is a money market fund a bank? They allow daily liquidations, but hold longer term maturity government securities. These are normally pretty liquid, so selling them in a hurry isn’t usually a big problem. Except sometimes it is, like when the Reserve Primary Fund “broke the buck” during the financial crisis as investors withdrew their money in a panic.

Is an equity mutual fund a bank? They too hold long maturity securities, and let you withdraw on a daily basis. Sometimes, you get fire sales where too many investors withdraw their funds and securities have to get sold at a loss, leading to greater losses for other investors. Sounds a lot like a bank run to me.

Is a public company engaging in maturity transformation if they issue short term debt and plan to roll it over when it matures? By any definition, yes. How on earth do you plan to stop that? Will our hypothetical regulator look into every public company and make an exact determination on when their assets will mature in terms of cash flows before debt can be issued? If you were the regulator tasked with assessing whether, for instance, Caterpillar’s debt timing was matched up with its expected earnings on bulldozer sales, how on earth would you propose to answer that question in anything like a definitive manner?

More importantly, how would you feel if the government, in the interest of banning maturity transformation, insisted that your credit card debt couldn’t be carried on an ongoing basis, but must immediately be paid off at the end of the first month? Also, you can’t roll over your credit card debt to a new debt with someone else either for the same reason. Can’t have any of that dangerous maturity transformation going on, old chap! I’m sure you’ll understand.

Or if you want to be pedantic, is a company engaged in dividend-smoothing also doing maturity transformation? They take lumpy and uncertain earnings and turn them into predictable, smoothed dividend payments. At least until the earnings don’t materialize, and the stock price drops. The maturity of the earnings, which is their asset, differs from the maturity of the equity dividends, which is a kind of liability. Admittedly shareholders can’t run on their own company, but the maturity transformation aspect itself is indeed occurring.

And all this is without even considering the rich variety of shadow-banking-type institutions that might pop up to offer securities that function like demand deposits in the old-fashioned sense. If you think that banning drugs is hard because there is such a demand for marijuana, wait until you see how large the demand for liquid short-term deposits is. There will always be creative ways to synthesize these, and regulators will almost certainly be playing catchup.

The point is, “ban maturity transformation” is a phrase that is very easy to say but much harder to either define the limits of, or to actually implement. And the difficulty has only gotten bigger as financial markets have gotten more complex.

Now, you might also look at some of the examples above and think that even if harm is caused, the harm isn’t sufficiently big as to actually ban the practice. And you’d almost certainly be right.

But this is exactly the point. It relates to another part of the analysis which mainstream economics offers, but fans of banning maturity transformation seem to not discuss as much. As economists are fond of pointing out, banning something is a form of a tax. It just happens to be an infinitely large tax (assuming the ban can be fully implemented). And viewed from this perspective, what we’re really saying is that maturity transformation imposes negative externalities. Which is absolutely true.

But the standard economics answer is that of a Pivogian tax on the externality. In other words, you want to set the level of the tax to offset the amount of the negative externality. If you impose a tax that is too high, you can end up doing more damage than the externality itself. Why would you suppose that the size of the negative externality from maturity transformation is so large that the practice has to be banned altogther? Why not just tax maturity transformation, perhaps with the tax reflecting the expected social cost of runs occurring in that particular situation? I’m fully on board with that. And indeed, deposit insurance schemes typically impose charges on the banks being guaranteed, with these charges acting in a similar manner to such a tax. But those who advocate a total ban need to explain why an infinitely large tax is the answer. Which, when phrased that way, becomes far less obvious.

The idea of a Pigovian tax also captures another aspect which is worth pondering. If financial crises are on the costs side of the ledger for maturity transformation, what is on the benefits side? In other words, what exactly is being lost if the tax on maturity transformation is set too high?

The answer, from standard economics, is otherwise productive investment. If depositors don’t like lending out for three years at a time, because they might have liquidity needs, then interest rates on three year loans will be substantially higher, and lots of otherwise productive investment projects won’t be undertaken. This will almost certainly be harmful to the economy. The Austrians sometimes have a habit of just waving this away, sometimes with the implication that the economy will just adjust, or that these projects shouldn’t have been undertaken in the first place. Perhaps. But there is definitely a wide range of evidence, much of it coming in the wake of the financial crises, that when banks reduce their available financing, firm investment declines. See here, or here, or here, or here, or just look up “credit constraints and firm investment” in Google scholar.

In other words, you would be very ill-advised to simply shrug off the potential loss in investment and output that might result from banning maturity transformation without considering seriously how costly it would actually be.

How large would this be, exactly? How valuable is all the marginal investment? And how much would investors respond to the higher interest rates by lending more? These are all great questions, and I don’t claim to know the answer to them. I also don’t think the economics profession can exactly claim to know the answer to them, though some people are studying this stuff.

There’s also the additional question of whether eliminating bank runs on its own would eliminate the periodic market crashes that we observe. It would definitely eliminate some of them. But all of them, or even most of them? As I’ve written about before, I think there’s a pretty strong case that the answer here is “no”.

But my real point is the following.

At this stage of the argument, we’re no longer debating reactionary versus progressive economics, or even really Austrian versus mainstream economics.

We’re just debating economics. We’re debating things at a level that is thoroughly within the economics mainstream, and for which mainstream economic thinking has lots of insightful points to make. What problems exist in the economy, and what are the costs and benefits of each potential solution?

Which is as it should be.

There are great economics articles written by reactionaries. There are even some covering areas which mainstream economics doesn’t traditionally focus on well (which perhaps was all my friend really meant, in which case I hope he’ll forgive this needlessly long and pedantic response).

For such articles, I offer the economist’s high praise.

They are great economics and other economists should take note.

Liked it? Take a second to support Social Matter on Patreon!
View All

16 Comments

  1. Jason Richwine’s dissertation would be a good example of reactionary economics.

    Reply

  2. Because you brought up deposit insurance specifically: Calomiris argues* (convincingly in my opinion) that the Diamond and Dybvig model is incorrect (not an accurate representation of reality). He instead proposed an alternative theory based on information asymmetry. Banks may or may not be _insolvent_ (as opposed to illiquid). Some of their customers have better information about this than others and they will preferentially withdraw their deposits (and other loans) before the rest. The wider client group can then use this as a hint that the bank balance sheet may not be as healthy as assumed. I highly recommend reading his book for more details.

    * See: U.S. Bank Deregulation in Historical Perspective by Charles W. Calomiris https://www.amazon.com/dp/0521583624/ref=cm_sw_r_tw_dp_U_x_PnouAb2NVPN3H

    Reply

    1. William Fitzgerald January 7, 2018 at 10:06 am

      Interesting. I hadn’t read that model, but it makes sense – certainly the line between banks failing from insolvency and failing from illiquidity is highly unclear. Fears of insolvency would suffice to create a bank run which finishes the bank, whether or not they actually were insolvent.

      Which, incidentally, is why everyone who suffered a bank run, from Enron to Lehmann, claimed that if they’d only just been extended a little more credit, it would have all turned out fine. That the claims are obviously self-serving doesn’t make them any easier to evaluate.

      Reply

      1. The argument that Calomiris put forth there is that illiquidity isn’t really a good reason for a bank run. To begin with, a bank has the option of getting short-term loans from other banks. If the bank really is solvent, why would the other banks refuse?

        Even if all the banks are getting stressed for some reason, there are still private alternatives that can prevent bank runs. The answer today would be mass securitization. The example that Calomiris pulled up from history was that banks would suspend convertibility (i.e. ban people from withdrawing from their accounts) and issue transferrable securities instead. This is actually a fairly attractive idea for customers. To begin with, the suspension can be combined with higher interest rates to compensate depositors for the temporary loss of liquidity. Depositors still have the option of selling their deposits to a third-part for immediate cash. Finally the alternative to suspension would likely be bankruptcy which was likely to be even worse. Of course convertibility suspensions make little sense in a world with the FDIC.

        What I find interesting about Calomiris’ model is that bank runs serve a useful economic purpose: they incentivize continuous monitoring of the bank’s balance sheet. If it starts to deteriorate, the bank is forced to clean things up or shut down. In contrast, if there was no MT and most deposits were long-term, the bank’s assets could continue to deteriorate and lead to much higher eventual losses.

        Reply

  3. It’s easier to tear down. It’s harder to build up. I applaud the effort.

    Reply

  4. *And governments can provide this guarantee in a way that private providers can’t, because the government has a firm ability to pay through its powers of taxation.*

    Sorta. But that would not be enough. The power comes from the ability to print money through selling debt, much in the same way a company does but with much lower yields and longer durations.

    good article

    Reply

  5. “f depositors don’t like lending out for three years at a time, because they might have liquidity needs, then interest rates on three year loans will be substantially higher, and lots of otherwise productive investment projects won’t be undertaken. This will almost certainly be harmful to the economy. The Austrians sometimes have a habit of just waving this away, sometimes with the implication that the economy will just adjust, or that these projects shouldn’t have been undertaken in the first place.”

    What you call ‘hand waving’ is the entire point of the Austrian theory. If the rate of interest in the absence of maturity transformation is 7% and the rate with maturity transformation is 3% then every single investment that is judged profitable only because the interest rate on a loan is 3% rather than 7% is a malinvestment. These malinvestments kick off the business cycle which makes a run on the banks inevitable. Deposit insurance cannot solve this problem; only printing a ton of money can solve this problem. This is why in reality, the only ‘stable’ fractional-reserve banking systems that have ever existed have a central bank and a fiat currency. However, that creates other problems.

    The article up to that point was pretty good. Not sure I agree, but a case well made for a debatable point. You should rewrite the article after reading ‘The Theory of Money and Credit’.

    Reply

    1. William Fitzgerald January 7, 2018 at 10:15 am

      I take your point that it’s not hand waving. But I feel sometimes that Austrians are too quick to dismiss the projects created under low interest rate regimes as “malinvestment”. What if they alternative is unemployment? I get that these will be the most marginal projects, but I wasn’t sold on the idea that this necessarily means they shouldn’t have been undertaken.

      As for this creating the business cycle – I’d believe that it might contribute to it, certainly if it actually leads to a banking crisis. But there seem to be business cycles in periods of history that lacked modern banking, and booms and crashes in experimental settings that lack anything even remotely equivalent. So at a minimum, there’s more going on in business cycles than just banking.

      Also, you said that deposit insurance couldn’t solve the problem. How come? I didn’t follow that part.

      That said, the Theory of Money and Credit is definitely on my future reading list.

      Reply

      1. Michael Rothblatt January 7, 2018 at 3:42 pm

        >What if they alternative is unemployment?

        That’s exactly what is the problem with liberals banning servitude. The solution is to bring back servitude, not make-work projects.

        >But there seem to be business cycles in periods of history that lacked modern banking

        Indeed there were.

        Reply

      2. “Also, you said that deposit insurance couldn’t solve the problem. How come?”

        Deposit insurance stops a run on the bank if the problem is unfounded rumors of bank insolvency. It cannot stop a run on the bank if the problem is well-founded rumours of bank insolvency as a consequence of providing loans for projects that cannot be completed because there are not enough actual saved resources to complete them.

        “What if they alternative is unemployment?”

        I think Moldbug dealt with this very well in one of his later posts. If your goal is employment as an end in itself, then just employ people directly. Build cool stuff like the Hoover dam. Get people to produce things of lasting historical value like slave interviews. Pay people to visit the elderly etc. Overheating the financial system is a terrible way of boosting employment.

        Reply

    2. Michael Rothblatt January 7, 2018 at 3:29 pm

      >However, that creates other problems.

      The problem with fiat money from reactionary perspective is that it creates a consumerist culture. Having inflationary money cumulatively increases the time preference of all. Thus, fiat money in itself is tantamount to decivilization.

      Reply

      1. I would say that monetary inflation is tantamount to decivilization. In principle, fiat money is the optimal currency system; the problem is the uses it is put to, which are probably inevitable.

        I assume you have read Hulsman’s ‘The Ethics of Money Production’, but everyone else here should read it too. It’s a shame that the outright imbecility of movement Libertarianism and the manifold nonsense that Murray Rothbard toyed with at various points has discredited the good parts of the Austrian Economics tradition for many Reactionaries.

        Reply

  6. There must be something really strange with your understanding. Avoiding bank runs was never a focus of ABCT.

    Why did you select this problem to “solve”?

    Reply

  7. “mainstream economists don’t consider enough the idea of simply banning maturity transformation”

    You can’t just “ban it.” You need an enforcement mechanism. The prudential regulation of banks in a system that banned maturity transformation would look exactly like the system we have now with deposit insurance.

    Reply

  8. Moldbug just endorsing Austrian economics was disappointing, because it did not deal with the more profound critique that economics is not in any way a science, if you take physics as your “gold standard.” Unlike physics, no economist can make consistently accurate prediction of the future. Just as neo-classical and Keynesians did not predict the Global Financial Crisis, the Austrians have been predicting that money printing will lead to inflation and a new crisis for TEN YEARS now, and none of it ever happens.

    Reply

    1. Michael Rothblatt January 10, 2018 at 1:33 pm

      >economics is not in any way a science

      Economics indeed isn’t science. Not in the way physics, chemistry or biology are.

      >if you take physics as your “gold standard.”

      Physics is, as a matter of fact, less “truthful” than economics. Economics is right there with mathematics, geometry, and logic.

      >Austrians have been predicting

      Which Austrians are you talking about? Pop-libertarianism isn’t Austrian Economics, you know. I know of exactly one (1) actual Austrian economist who predicted hyper-inflation. The rest are either pop-libertarians, or goldbugs who have interest to push the collapsitarian narrative so that they could sell you gold.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *