When people talk about the financial services industry and subsidies, they’re usually thinking of implicit subsidies the industry receives — you can cast the bankruptcy code, the tax treatment of interest, and even the existence of central banks as an implicit finance industry subsidy. But it’s a tricky topic: if you subsidize a middleman, are you sure you’re subsidizing the middleman and not the customers? If everybody has access to the central bank when times are tough, no one person with access gets a net advantage; the lender of last resort just increases everybody’s risk tolerance.

This middleman characteristic means that if finance as an industry gets a subsidy, it’s mostly passed on to either savers or borrowers. The lender of last resort, for example, is a borrower subsidy most of the time, but a saver subsidy during a crisis.

But there’s another interesting way finance and subsidies interact: there are some activities that the industry itself tends to subsidize. A large financial services industry can only thrive by aggressively extracting capital from industries with poor economics, and investing in industries with better economics. So the size of the financial sector is, among other things, a measure of the invisible productivity subsidy from asset allocation.

The cartoon description of finance is that it does this by, well, financing: if investors are willing to back a semiconductor foundry but not a steel mill, that means more semiconductor foundries get built. But as many critics point out, most transactions do not constitute flows from investors into businesses; they’re flows from investors to other investors. If a high-growth company IPOs and never raises capital again, Big Finance can’t exactly pat itself on the back for growth that the company funded mostly through internal operations.

Or can it?

Secondary transactions — shareholders selling to new shareholders, traders whipping around treasuries, etc. — have second-order effects.

Returning to the steel mill example: a steel company might be profitable today and could wonder what they can do to be as valuable as possible ten years from now. They look at their stock price, and see something interesting: they’re trading at less than book value, and so are their competitors. A billion dollars worth of steel-manufacturing assets transform into less than a billion dollars worth of steel company stock!

They could completely ignore this and go about their merry way, but the transactional nature of finance means that everybody who gets mad can make a trade that gets them even. If the steel company feels that its prospects are misjudged by the market, the smart thing to do isn’t to just keep making steel — it’s to make steel, and use the proceeds to buy back stock. If you can reduce your share count while paying eighty cents on the dollar, the shareholders who stick around get rich.

The net effect of this is, of course that the steel company — which is spending its money buying stock on the open market — is not spending that money on new steel-producing assets. The invisible hand is more invisible than usual: a pension fund in Buenos Aires offloading stock to a mutual fund in Paris can cause a steel company in Cleveland to rethink its priorities.

The same dynamic, inverted, shows up for high-growth companies. Generally, they find that the best way to attract top employees is to offer them equity. But when your employees have most of their net worth in the company, stock price declines become a problem. So, merely as a matter of motivation, the company has to pay attention to what the stock market wants — even if they’re never going to raise capital again.

(This doesn’t mean companies should be slaves to the market, of course. A heterodox CEO who knows something the investing public doesn’t should act on that knowledge, even if it pushes the stock price down in the short term. Crucially, even if investors don’t buy into the logic, employees probably do — the best ones even more so. It’s hard to take a hit to your net worth, but if your perspective is that short-sighted idiots just cost you a ton of money, you’ll be twice as motivated to show them up by proving them wrong.)

Why Care?

Finance apologetics is tricky. I am, as we like to say, talking my own book here. However, it’s important to talk about the productivity effects of finance because the productivity of finance is so hard to measure. Suppose the industry mints a bunch of money some year. Is this because:

a) It created a lot of wealth?

b) It got better at extracting economic rents?

c) It found a new form of risk premium, and earned revenue now in exchange for paying it out later?

Nobody knows. A priori, everyone in the industry can argue a) and every skeptic can argue b) or c). There are a lot of verbally-adept people in this world; they’re disproportionately likely to work for banks or work as journalists, so the public debate about finance is a sort of fratricidal intra-elite class war between two factions who’ve hated each other since prep school.

But this trilemma applies to other industries, too. Cigarettes, as it turns out, are a novel way to collect superior short-term IRRs in exchange for the risk of catastrophically expensive legal settlements, or desperate investments in JUUL. Some manufacturing companies collect a risk premium against giant EPA fines. Retailers surf trends, until they can’t.

The marketing and advertising industry might appear to play a similar role to finance: it doesn’t produce, but makes other fields more productive. However, this is hard to argue: finance as a share of GDP has risen rapidly over the last few decades; advertising as a share of GDP has been surprisingly stable. If finance is growing because, in a mature economy, the important problem to solve is figuring out what to invest in, you’d think this same abundance would raise the amount of money invested in ads. Not so, at least not yet. If anything, ads as a share of GDP should be going up, because digital marketing turns some parts of retail into ads instead. A comparison shopping engine is basically a mall made out of pixels, but its revenues are advertising while the mall’s are not.

Another reason for apologetics is the usual one: the critics may have a point. Finance absorbs a lot of human capital. (Mine, for example.) Some of this capital could be put to work in other, more productive ways. But the industry employs broad swathes of people who, if they weren’t at a bank, would not be curing diseases and building awesome rockets; they’re people who would be working in some other white-collar service job. If a side effect of the growth of finance is that the rest of the economy hums on a little quicker, a few fewer rocket scientists might net out to a winning trade.