Is Finance Too Big? Some Regulation Basics [Econ for Poets]

Allow me to come at the question from this angle: We often hear it said that the finance industry has gotten too big. In other words, we have too much financial services. But if you’ve had introductory economics, your ears should prick up at this claim. What would it mean to say, and how could it be the case, that an industry had gotten too big?

Look at it this way: What if somebody told you that the shoe industry had gotten too big? If you think about it, this claim doesn’t actually make much sense. The shoe industry is simply supplying as many shoes to the market as we shoe-wearers demand at the price that covers its costs plus a small allowance for profit. In other words, it’s weird to say that the shoe industry is too large, because its size is determined by what we (society) demand. Sure, maybe people should be less vain and more willing to wear ratty shoes for years on end, in order to save money that they could use to solve global poverty. That is actually (I’m serious) a morally reasonable claim. But in that case, your issue is with society and its mores, not with the shoe industry. Remember back to the post on sin taxes: In a competitive market, unless there are externalities, the price of a good is a measure of its cost to society as a whole. Every time an individual buys a pair of shoes, then, he is himself covering the social costs of the production of those shoes—and he’s saying that he’s more interested in spending those X dollars on those shoes than on any other thing. So the shoe market as a whole is just the aggregation of individual transactions in which people say, essentially, “These shoes are worth more than anything else I might the money toward; and I’m willing to bear the full cost to society to get them.” The abstract term “the shoe market” actually just refers to that thing happening billions of times. So, if we’re non-judgmental, and tolerant and accepting of individuals’ preferences and purchase decisions, then we have to admit that, yeah, the shoe market is about the size that it should be—it must be.

Is finance different? “Yes, of course,” everyone says. And everyone is right. But they’re usually right for the wrong reasons. People say things like, “Finance is not real work—they don’t produce real things, like in manufacturing, but they still come away with all the money.” Sure, but same goes for your general practitioner doctor—he just produces advice, and puts symbols on papers so you can pick up prescriptions that machines have made, and claims a hefty salary. Is that itself a reason to think you have too much of your general practitioner doctor? No. Indeed, in modern America, must of us are employed in services, as opposed to the production of physical commodities. “Right,” people say, “but medical services produce value for people directly, but finance just moves the money around. It doesn’t produce new wealth for society; it just captures it.” And while this might, debatably, be true for certain kinds of high-frequency trading, it’s not true of finance broadly. (Brief explanation: There’s a strong correlation, and a logical causal link, between the development of a society’s finance industry and its broader level of economic growth. One of the biggest determinants of whether a business succeeds or fails is its access to credit and its insurance against unpredictable risks. These both depend on the robustness of the financial services industry.) If you think that figuring out what firms and projects could best make use of investments is easy, trivial business, then you should probably already be a billionaire right now. So a priori, there’s no reason to think that finance is fundamentally different from other services.

So how could the financial services industry be “too big”? Well, interestingly, the financial services industry could be too big for the same reasons any other industry could become too big—for example, if it is directly or indirectly publicly subsidized. Remember back to the sin taxes post, where we saw that if a good had negative ‘externalities’ too much of it would be produced. Well, goods and services will get overproduced when they are publicly subsidized for the same basic reason—individuals are not bearing the full costs of their purchases. (Unless the public subsidy has been put in place to directly balance the goods’ positive externalities—but this is for a different post.) I.e., if the government were to pay some of the cost of every shoe purchased in America (which might happen, say, in an alternative universe in which shoe-wearing became a hot spot in the culture war, such that one party, eager to flare the culture war for its own fundraising purposes, decided that shoes qualified as something to be covered by health insurance), we would all go out and buy more shoes than were really ‘worth it’ for society as a whole, because each individual would be getting shoes for less than they cost society as a whole.

So: Is the finance industry publicly subsidized? Yes, it is! But how? The answer is a term that is very familiar: Too Big to Fail. As we saw in that little financial crisis that you may have heard something about, when a so-called “systemically important” financial institution looks like it is about to fail, the government will step in and use public funds to save it. Those funds, and their implicit guarantee, act as subsidies for the specific “systemically important” financial institutions that are protected by them and for the system as a whole. When the government, in times of crisis, lends to these institutions at below-market rates, and buys their toxic assets at above-market rates, that sort of amounts to a more publicly palatable way of just giving money away to them. And the knowledge that the government will save them in a time of crisis it easier for these “systemically important” financial institutions to get more, cheaper credit (loans) from the private sector the rest of the time, because the knowledge that these firms won’t be allowed to fail means that the creditors to “systemically important” financial institutions don’t need to worry about the risk that they’ll default.

Now, here’s an interesting thing: Right now, the conventional wisdom is that “Too Big to Fail”—the policy—is necessary and unavoidable and good. That is, the CW is that it is just factually true that these systemically important financial institutions will bring down the whole financial system, and thence the whole economy, is allowed to fail, so we have no choice but to bail them out, and there’s no way around that. The real problem, the CW continues, is not the policy, but “Too Big to Fail” the fact of life—i.e., the fact that these financial institutions have gotten big enough that the failure of just a few private firms could bring the whole economy down. The wicked irony, of course, is that once “Too Big to Fail” the fact of life exists, then “Too Big to Fail” the policy becomes necessary, and the policy itself helps perpetuate and exacerbate the fact of life. That is, as we noted above, the “systemically important” banks have an implicit subsidy and implicit guarantee that helps them get easier, cheaper access to credit than smaller, non-systemically important banks, which helps them outcompete these smaller banks, and gives every bank an incentive to become dangerously large. More, once banks have gotten dangerously large, giving them implicit TBTF protection, they actually have a publicly-subsidized incentive to make extra-risky investments and bets, because they get to claim the full upside of those risky investments, while taxpayers have to bear the downside costs if they go sour.

So, this together had led to a new conventional wisdom in the debate over “What do we do about this shit, post-financial crisis?” And this conventional wisdom is that trying to limit executive pay won’t do anything but satisfy our envy-lust; separating commercial banking from proprietary trading won’t really do the trick anymore; and just giving regulators more generic authority won’t really work either, because the whole thing about a financial crisis is that it comes from risks that we don’t see. Rather, real financial reform needs to start with breaking up the big banks, so that no bank is actually, in fact, TBTF. Once this happens—that is, once no bank or small group of a few banks is so large that its failure poses a risk to society as a whole—no bank will have an implicit government guarantee. So, no banks will have such an accordant specially, unfairly easy time getting capital (which should slow the consolidation of the financial system). And, lenders to all financial institutions, knowing that the government won’t bail them out, will be more circumspect in examining these financial institutions’ risk—which should, combined with their own financial interests, put pressure on banks’ managers to do better and more circumspect risk management.

In short, the key to fixing the financial system, the new CW goes—and I think there’s a lot of truth to this CW—is to actually allow market pressures and incentives to work on the financial services industry as they should in a real market.


So now that I’ve introduced you to the conventional wisdom, let me tell you about the subversive, contrary-the-conventional wisdom piece I read that forced me to revisit the conventional wisdom in the first place. In Dealbook, Richard E. Farley writes,

“But would we be better off if no bank were too big to fail? And what might happen if all the banks were too small to save?

One thing we know for certain: banks will continue to make loans that will not be repaid, none of which, when initially made, were thought to be bad loans. Many of these bad loans will be discovered to be unsound at the same time across many banks. When the scale of loans going bad all at once is large enough, you will have a banking crisis.”

Read the whole thing. But here’s the basic idea: All financial crises happen from widespread defaults on loans that all banks are making, which result from unpredictable shifts in the real economy. So, breaking up big banks doesn’t actually reduce systemic risk—it just redistributes it into more separate, but equally interlocked, balance sheets. So the size of the banks per se doesn’t really matter. There are going to be crises anyways, because sometimes large numbers of people just get things wrong and don’t see changes in the economy coming. Given this, a TBTF policy is the only thing that can stop ‘bank runs’ (think of the bank run in It’s a Wonderful Life—except this isn’t individuals demanding their deposits back [a phenomenon prevented with the genesis of the FDIC], but banks and hedge funds etc. demanding their money from each other, cutting off overnight loans, etc.). With a few large, systemically dominant banks, it is easier for the government to restore confidence in the system—it was able to summon the “big 9” to the New York Fed and get them to work together on a bailout policy, but it would be harder to assemble hundreds of smaller, less reputable bank managers.

Is this argument correct? Yes and no—I’m not really sure. I ran the idea by a few smart friends on Facebook, and here’s my basic sense now. If you read closely, you’ll notice that Farley ignores the main, big argument in favor of breaking up the big banks, in that he implicitly assumes that a hypothetical alternative group of smaller banks would, together, make the same investments our large, ‘systemically important’ banks are making right now. And, according to our logic above, that’s not true. The big banks make riskier bets because they know they’ll be bailed out by the government if they things go seriously wrong. Smaller banks won’t have this guarantee, so they’ll be more circumspect. So that part of Farley’s argument seems to fail.

But, still, if there are weird enough changes in the real economy, a big financial crisis still could happen, even in a world with very small banks. So what are our options given that? I see two: One is to say that, whatever its costs, TBTF is okay, because it is the only way to stop bank runs during a time of crisis. The other is to say that with a few more tweaks to the system—say, for example, forcing banks to hold more equity and lower levels of leverage, so they can bear larger losses without going under—we could make financial crises so surpassingly rare, and less impactful on the real economy, as to make it unnecessary.


One thought on “Is Finance Too Big? Some Regulation Basics [Econ for Poets]

  1. This is probably wrong, but I’m going to post it anyway. Here’s another reason the anti-CW perspective might oversimplify things: correlated behavior. Why? It’s a lot easier for the big 9 to track each other’s actions than any one of 500 independent small banks with limited resources to track the other 499. That is, the strategies pursued by any one of the 500 will be more independent of those pursued by the other 499 than the strategies pursued by any one of the big 9 are of those pursued by the other 8. Correlation in strategy produces volatility. When things work, they really work. When they fail, they really fail. Busts (and booms) are significantly more likely when there is less strategic heterogeneity in a system.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s