Lord Adair Turner, who was the chairman of the UK Financial Services Authority from 2008-2013, gave a brief but hard-hitting speech last week, where he claimed that university textbooks were teaching a ‘mythological’ story about what banks do. Speaking on a panel at the INET “Human After All” conference in Toronto, he made a number of strong points:
- That much activity in the financial sector is not useful to the economy, and that we should discard the pre-crisis belief, held by most economists, that a bigger financial system is always beneficial for the economy
- That the idea that banks take money from savers and lend it to borrowers is wrong. In fact, banks create new money when they lend.
- That the idea that the function of modern banks is to invest in businesses is ‘mythological’. In reality, lending to businesses is often below 15% of the loans that banks make. Instead, most of the money they create (by lending) goes into property bubbles and financial markets.
The speech is well worth watching in full, but if you’re short of time, you can get the highlights by reading the bits in bold below.
“Good morning everybody.
It’s a striking fact that over the last 50 years, finance got much bigger across most advanced economies. The broad figures for the US set out, for instance, in Robin Greenwood and David Scharfstein’s paper, was that in 1950 the US financial system accounted for about 2.5% of US GDP. By the mid-1970s about 4.5% and on the eve of the crisis about 8% of GDP. And as Paul [Martin, former Canadian prime minister] on some other measures such as gross operating surplus or equity value, much higher percentages than that. Andy Haldane’s figures illustrate the same figures for the UK.
In addition, the financial system has drawn to itself highly talented people but paid them an enormous amount of money – even more than you can explain by the amount of talent. So the analysis by Thomas Philippon and Ariell Reshef shows an extraordinary excess over what you would expect to get for the skills level of the people in that industry; an excess which has varied over time. It was about 70% above other sectors of the economy in the 1920s; it then fell to about 0% in the 1950s and 60s, but by the eve of the crisis it was back at 70%.
So we have a large and growing sector of the economy, paying its participants – its producers – a large amount of money. And we have, as Joseph [Stiglitz] said, to ask questions about that. We can’t simply accept that this is ok in the same way that we would if say, the restaurant industry grew. If the restaurant industry grew from 2% to 8% of the economy, we would say, well that’s because people like buying restaurant meals. But nobody gets up in the morning and says, “Ooh, what shall I do today? I think I’ll go and consume some financial services”. Well, there may be some people in this room who do that, but most normal people don’t do that. It’s good if it’s performing good functions vis-a-vis the economy. It’s not if it is not.
So we have to ask searching questions about why finance got so big, what it is doing, and whether that is valuable. Now again Greenwood and Scharfstein have done a very valuable exercise in breaking down the growth of the US financial system into what constitutes it. And it’s useful to divide what’s occurred into two elements: 1) a change in what the financial system does vis-a-vis the rest of the economy, i.e. actual services to the rest of the economy, and 2) what it actually do itself.
If we focus on what it does vis-a-vis the rest of the economy, you find some things which have grown, but no more than you might expect, like insurance – insurance services have grown because we’ve got more things to insure. But actually the growth of finance vis-a-vis the rest of the economy is dominated by two things.
One is credit. The financial system makes money out of the net interest margin on credit and the fees related to the credit origination, because we borrow more money. In the US in 1950, private credit-to-GDP was 50%. It rises relentlessly to 170% of GDP by the eve of the crisis. Because we borrow more money, the financial system makes more money out of us.
The other thing, which is of course linked to this – indeed it’s almost just the flipside of this on the other side of the balance sheet – the industry makes much more money out of asset management in all its many characters: mutual fund fees, private equity fees, hedge fund fees and brokerage commissions and various forms of trading. And of course if there’s more credit there’s also something more on the assets side which has to be managed.
So there are more services to the real economy. But the other striking feature of the growth of the size of the financial system is that it’s doing an enormous amount of things that it’s doing with itself. If you look at a bank balance sheet from 1950, you can understand what’s going on. You will find that there are loans to the corporate sector, or deposits from; there are loans to the household sector; there’ll be some liquid assets held in the form of government bonds, but the bit which relates to the real economy will be the majority of the balance sheet. You can understand it. If you look at a major trading bank [balance sheet] now – the Goldman Sachs or the Lehman Brothers (as it was) or the J. P. Morgans or the Barclays – the vast majority of the balance sheet has to do with trading vis-a-vis the rest of the financial system. It has to do with prime-brokerage relationships; it has to do with enormous amounts of interbank placements.
And one of the striking features of what’s occurred is this amazing increase in intra-financial intensity. In trading volumes – trading volumes have increased dramatically relative to the real economic activities to which they relate, and we’ve created an entire infrastructure of contracts such as derivatives and structured credit which didn’t exist before.
Now the pre-crisis orthodoxy was that both of these developments – the rising intensity vis-a-vis the real economy and the rising intensity within the system – were good. The rising intensity vis-a-vis the real economy was good because [a rising ratio of] credit-to-GDP was a positive thing; there’s a lot of finance theory that suggests that, and argues that for instance India or Bolivia don’t have enough credit to GDP. There was a broad supposition in favour of financial deepening.
As for the increase in intra-financial intensity, that was justified essentially – as Joseph Stiglitz has suggested already – as a sort of hand-me-down from the nirvana of bliss of the Arrow-Debreu general equilibrium analysis. If only we all just create as many contracts as possible, complete all markets, and trade continually between all the different agents, we will arrive at the Pareto-efficient maximum possibility of human welfare. And that belief was embedded in a set of language (and I think this is an important point for us generally in INET) a set of language which rhetorically constrains the debate. Language such as “market completion”: well you wouldn’t want to be against market completion; things must be better if they’re more complete, not less complete. “Liquidity”: liquidity is always better. “Price discovery”: how could you possibly be against discovering things? Or “credit”, which Mr and Mrs Ordinary Joe want to build their business?
The crisis showed that that wasn’t true; that we cannot have confidence that this system was generating – as was asserted – both greater efficiency and greater stability.
I think there are two major issues we have to ask about the financial system. One is simply about efficiency, and value for money, and we can focus that question on all this amazing intra-financial system activity – this trading activity, these contracts, this more complicated (as Axel Leijonhufvud said) web of relationships within the financial system. The axiomatic assumption of neo-classical economics and the pre-crisis orthodoxy is that it must be adding value because it is completing markets. But Joseph Stiglitz’s work, above all, has helped us understand that that isn’t necessarily true; that more trading under some circumstances can be excessive because trading in financial instruments is not trading between two people who have different consumer preferences or production possibilities; it’s trading between different people with different points of view over the state of an uncertain future and that is different. More trading can be harmful.
It’s also not true that more market completion is necessarily good. You create a CDS which is capable of hedging a risk, but the very act of creating such a CDS can enable other people to bet in a way that can destabilize the economy.
So there is a possibility that we can generate within the financial system an essentially casino activity. Of course, it raises the question of, why does that casino activity exist? Why does it survive? If this is a zero-sum trading game between people, why don’t the systematic losers leave the game, and why does it matter to the rest of us if these guys care to go off into a casino and bet against one another?
Well it matters because that casino is to an extent kept going by a set of rents that arise in the retail system and in the relationships between the retail and the wholesale, the scale of the asset management fees, and the imperfections in the ability of the ordinary consumer to buy in an efficient fashion.
So I think there is a reasonable argument that we are not getting value for money; that we are getting a proliferation of activity which is performing the fundamental function of intermediation more expensively than it needs to do.
But it’s not the biggest issue. Suppose that 8% of GDP didn’t need to be 8%. Suppose we could have done this function for 6% of GDP? On average we’re 2% worse off.
But we’re 10% worse off, in all rich developed economies, compared with the trend because of the crisis of 2008. And what that tells us is that the much bigger issue is the instability issue. The value for money issue, the wasted activity, the activity which isn’t required, is an important issue, but it is not as important as the instability issue.
And I think it’s clear that greater financial intensity made the system more unstable but in a way that really poses very difficult issues at the interface between finance theory and macro-economics. Because I think that when we get to the driver of instability, the most important drivers are not actually the intra-financial system intensity. I think that’s important; I think intra-financial system intensity created what I call the credit cycle on steroids; a hardwiring of instability in the way that people like Markus Brunnermeier and Hyun Shin [of Princeton University] have described; a way whereby we use things like ‘value at risk’ and mark-to-market and secured financing in a way which made the system unstable. But while that inner-guts of the financial system, the web of contracts, is important, the even-more important driver of financial instability is simply rising leverage of a particular form and of a particular form on which finance theory and macroeconomics has been entirely silent, or mainly silent, or to the extent that it has not been silent, it has been wrong.
If you pick up most undergraduate textbooks (and we were talking earlier this morning about the role of what we have to do in teaching undergraduate and graduate economics) and you see how they describe the role of the banking system, they make two mistakes. First of all they describe a system which takes money from savers, and lends it to borrowers, failing to realise that the banking system creates credit, money and purchasing power ab inicio, de novo, and with an important role therefore within the economy.
But also, again and again, [the textbooks] say “Well what banks do is they take deposits from households and they lend money to businesses, making the capital allocation process between alternative capital investment.” As a description of what modern advanced economy banking systems do, this is completely mythological.
This is very well described in a forth-coming paper from [economics professors] Alan Taylor and Moritz Shularick, reflecting some INET-sponsored research. Basically in advanced economies, 80-85% of bank credit is not to do with funding capital investment. It is funding two things. It is funding either consumption: that may be valuable but we need to understand why that is and why funding consumer credit would be valuable and to what extent. And the vast majority of it funds real estate – either residential or commercial. Sometimes that is funding new real estate investment, which is a form of investment in the economy, but most of it is funding the purchase of already existing real estate. And it is fundamentally funding – in a credit form – a competition between ourselves for the ownership of locationally-specific urban land.
And that, I think, is absolutely fundamental to the instability cycle, because it is that competition for the ownership of existing real estate, and of location-specific land, which is in relatively fixed supply, that creates instability. Because if you put together the fact that the banking system that can create infinite quantities of credit and money and purchasing power, and that urban land which is locationally desirable is in fixed supply, whenever in economics you put together something which is infinite and something which is fixed, you get an inherently indeterminate price.
We have at the core of our financial system something which is bound to create inequality, bound to create credit and asset price cycles and booms, bound then to create crisis and debt-overhang and the balance sheet recessions about which Richard Koo has talked so importantly.
That really poses a fundamental challenge, because it means that we have a free market system which left to itself will create a problem, not just of a deadweight cost of too much intra-financial activity that we don’t need, but will also create too much of the wrong sort of debt – but with the irony that each one of those debts seen on its own looks perfectly socially useful: how could you deny that someone borrowing for a mortgage is socially useful? But the collective impact of it is something which is incredibly difficult for us to manage, and a fundamental problem at the interface between finance and macroeconomics.”