What does a bigger financial sector buy an economy?

The fee you never see

Look closely at a pension or retirement statement and you will find, usually in small print, a figure for costs: an annual charge, quoted as a fraction of a per cent of the money under management, deducted before the balance is struck. It is a small number, and it is easy to pass over. A charge of one per cent a year on a pot of savings sounds trivial next to the returns the fund reports, and the money leaves so quietly that most savers never register it at all. Yet that quiet clip is a slice of your retirement handed, year after year, to the financial sector for the work of managing, trading, and safeguarding your savings. Over a working life it compounds into a meaningful share of the final pot, and it is paid whether the fund does well or badly.

Where that pension money ultimately goes, who pays for old age, and why a fund can hold decade-long assets while standing ready to pay you next year, are questions this volume has already opened in the chapter on stock markets (the previous exchange chapter) and shares with the volume on work and the welfare state (Volume IV, on who pays for old age). This chapter follows the fee rather than the pension. It asks about the sector on the receiving end of that charge, and of a thousand charges like it: the banks, funds, insurers, brokers, and traders that together make up finance. That sector has grown far faster than the economy it serves, and the plainest way to ask what the growth is worth is to put its rising size beside the one measure of what it costs to do its core job.

A sector that outgrew its economy

Start with the size. The share of the American economy taken up by finance and insurance, measured as the sector’s value added against national output, was about 2.8 per cent in 1950. By the mid-two-thousands it had reached roughly 8 per cent, close to a tripling in sixty years, and the sector’s share of all corporate profits rose by even more, from something like a tenth in the mid-century decades to roughly a third at its peak just before 2008. A part of the economy that had been a modest utility became one of its largest single industries. This growth was not a smooth trend rising out of the 1950s. The economist Thomas Philippon, reconstructing the series back to the nineteenth century, found that finance had been large before, reaching an earlier peak of almost 6 per cent of national output during the Great Depression of the early 1930s, and was actually smaller as a share of the economy in 1980 than in the 1920s, before the long climb that followed. The modern financial sector is not the natural end-state of a maturing economy but the high point of a curve that has risen and fallen before.

The cost that did not fall

A tripling in size would be easy to explain if the sector had been getting steadily better and cheaper at its work, the way computing or telephony did over the same decades. To test that, Philippon asked a deliberately simple question: how much does it cost, each year, to move one dollar through the financial system from the people who save it to the people who use it. He measured the whole income of the finance industry, all its wages and profits together, and divided it by the total quantity of assets it was intermediating, the loans, bonds, shares, and deposits it stood between. The result is a unit cost, the price the economy pays for one dollar of financial intermediation held for one year. If finance were becoming more efficient, this cost should fall.

It has not. Across roughly a hundred and thirty years, from the 1880s to the present, the unit cost has stayed close to 2 cents on the dollar, wandering in a narrow band between about 1.5 and 2 per cent with an average near 1.9 per cent, and showing no downward trend. It costs about as much to move a dollar through the system today as it did in 1960, and about as much as it did around 1900. This is the finding that gives Philippon’s paper its title and its force: over a century in which the telephone, the computer, the electronic exchange, and the securitised market all arrived, and in which the volume of assets handled grew enormously, the measured price of the core service did not come down.

What the extra scale bought

Put the two figures side by side and the shape of the puzzle is clear. The sector roughly tripled its share of the economy, and its profits and its top salaries rose by more, while the measured cost of its core service held flat. Something was bought with all that extra scale, and the answer is that it was several different things at once, not all of the same worth. Robin Greenwood and David Scharfstein, tracing where the post-1980 growth actually went, found two engines behind it. One was the management of financial assets: as household wealth and share prices rose, so did the pool of money professionally managed for a fee, and the fees rode up with it. The other was the vast expansion of household credit, above all mortgages, which grew from under half of national output in 1980 to nearly all of it by 2007.

Some of what that scale bought is genuine service. A modern economy runs on far more credit reaching far more people than it did at mid-century, and lending to a young household with no collateral, or to a small firm with an untested idea, takes more screening and more risk-bearing per dollar than rolling over a blue-chip loan; Philippon himself notes that extending credit to harder-to-judge borrowers requires more intermediation, so part of finance’s growth reflects real work on assets that were genuinely harder to handle. Wider access to mortgages, insurance against more kinds of risk, and the pooling that lets your pension hold a sliver of a thousand companies are things the larger sector delivers and the smaller one could not. Some of it, on the other hand, is what the growth’s own students call rent: activity that swells the sector’s income without a matching gain for the people it serves. Greenwood and Scharfstein note that active asset management has stayed expensive while rarely beating a cheap index over time, so that much of the money paid for it buys the hope of outperformance rather than the fact, and draws more talent and resources into the industry than the service returns. The flat unit cost is the aggregate signature of that mixture: real services and unearned tolls, growing together, with no net cheapening to show for a century of technology.

Plumbing, or a toll on the pipes

This is the ground on which the sector’s defenders and its critics have long fought, usually by choosing one half of the picture and calling it the whole. To its defenders, finance is the plumbing of a modern economy, the system that carries savings to their most productive use, and a richer, more complex economy naturally needs more of it; the growth is the price of sophistication, and to begrudge it is to begrudge running water. To its critics, the sector is a toll booth on those pipes, a rent-extractor that has learned to take an ever-larger cut of the economy’s flow while adding little in return, and whose brightest talent and richest rewards are a drain on everything else. A former head of Britain’s financial regulator, Adair Turner, gave the critical view its sharpest modern phrase when he called swathes of the sector’s activity socially useless, a judgment quoted in the previous chapter on the stock market.

The measured record does not hand the argument to either side. What the record shows is a sector that became much larger and much more profitable without becoming measurably more efficient at moving a dollar, and whose growth bought a real expansion of credit and risk-sharing alongside a real expansion of fees for activity that does not clearly repay them. The plumbing is genuine and so is the toll, and they run through the same pipes, which is why a century of cheaper technology did not lower the bill. Whether the extra scale is worth its price is a question about the balance of those two, and it is not settled by the size of the sector alone, which is the mistake both poles make. The stable unit cost is the single most disciplining fact in the debate, because it means the burden of proof falls on the claim that the bigger sector is also the more efficient one.

Where the argument goes next

The bigger financial sector, then, is not simply what a wealthy economy grows into, nor is it a straightforward story of a service getting better and cheaper. It is a part of the economy that has roughly tripled its share since 1950 while the measured cost of its core task held near two cents on the dollar for a century, which means the extra scale bought a mixture the aggregate figure cannot separate: a real widening of credit and risk-sharing on one side, and a real growth of fees for activity that does not clearly repay them on the other. The fee clipped from your pension pays for both at once, and what it buys you depends on the balance between them, which is a question of degree rather than of plumbing against parasite.

There is a further cost that the flat unit cost does not capture, and it is the one savers feel most sharply. A larger, more interconnected financial sector is also a more fragile one, capable of turning its own troubles into a crisis the whole economy pays for, as it did when the mortgage machinery described here collapsed. Why those crises recur, how the system’s growth and its fragility are linked, and who ends up carrying the bill, is not settled here; it is the subject of the volume on what goes wrong (Volume V, on why booms and busts keep happening, and the crash of 2008). Nearer to hand, this volume turns from the sector that moves money to the newest market of all, the one that trades not in savings but in attention, and asks how the thing you give a free app became a product sold over your head. How attention became a commodity is the next chapter’s question.