After last week’s festival of secular stagnation (StagFest?), this week’s trend kicks off from Paul Krugman’s post wondering why the Swedish central bank is raising rates. Simon Wren-Lewis gets into it more. The Riksbank is worried about property prices, and the banks that love them. They’re afraid that a property bubble might break out, and hope that cranking up interest rates for everybody will stop it.
At the same time, the Reserve Bank of New Zealand and the Bank of England are taking a different approach. The New Zealanders have imposed a regulatory limit on big mortgages, defined by loan-to-value. They’re not totally banned, but each bank has a limit in how many they can give out above a given LTV. The Bank of England has vetoed any more funding for property loans under the joint Bank-Treasury Funding for Lending scheme. Meanwhile, Germany may be going to impose rent controls, which is important if you think property is worth the net present value of its rent.
So what’s up here? Consider the case where the market for property is rocketing while the rest of the economy is far from full employment on some metric like…uh…employment. The interest rate that would be appropriate for the property sector is different from that in the wider economy. Theoretically, the market for capital ought to smooth this out but observably it doesn’t happen. (Like the difference in interest rates within the Eurozone between German and Italian SMBs.) Lowering interest rates for the wider economy will cause even more trouble in housing, while upping them to stop the insanity will impoverish still more people. We could call this the UK case, as it happens in Britain all the time. Another example would be a strong, indeed overheating, macro-economy and a housing market that is already far enough along the Minsky scale that pulling up the interest rate to slow the wider economy will cause a financial crisis starting in the property market.
These are, in a sense, the same problem. What’s happened in both is that monetary policy control has been lost. We can look at this in two ways – either it’s impossible to use monetary policy effectively, because the consequences are so bad, and therefore it’s no longer a practical instrument of control, or else the central bank no longer controls monetary policy itself. To see how this might happen, think of the second case. Whether the crisis is because a marginal buyer can’t afford to buy at the new interest rate, or because the policy change is taken as a signal and sellers unload, asset prices dive and financial institutions end up in trouble. The magnitude of the increase in market interest rates, or the volume of credit provided in the market, is decoupled from the policy input. This is what it means to lose control.
In some circumstances it might even be possible to have a perverse response to control – Mike Konczal explains, and foreshadows a point that will be important later.
It’s worth thinking about the relationship of bubbles and monetary policy for a moment. A good working definition of a bubble is a situation where the capital gain an investor expects over their operating horizon is much greater than the potential change in interest rates in that time.
J. K. Galbraith made this point about the Wall Street crash of 1929 in The Great Crash. The typical speculator of the time bought on margin, borrowing from a stockbroker, who themselves borrowed from the world capital market. Brokers’ loans in the summer of 1929 were repayable on call and attracted a 20% interest rate, but even with these terms, people made money fast, as long as the market went up. Galbraith pointed out that, therefore, no movement in interest rates the Federal Reserve could bring about would stop the bubble.
He also pointed to the role of foreign companies and wealthy individuals who invested directly in brokers’ loans, attracted by the high rates and call terms, who unlike banks didn’t draw on Fed funds and therefore didn’t care about its rates. But this is beside the point. If the bubble is going up fast enough, the change in interest rates required to stop it may be either unachievable because the policy instruments can’t deliver it, or else it may be so large that the consequences for the wider economy are unacceptable.
Minsky’s three phases are interesting here. In phase one, hedge finance, the cash flows from a typical asset pay the interest and principal and something more. In phase two, speculative finance, they pay the interest, and the resale value of the asset has to go up to make the deal work. In phase three, Ponzi finance, they don’t even pay the interest. Another way of saying this is that the interest rate is now irrelevant. It’s win or Sing Sing.
A further point here is that the market interest rate determines the boundaries between the phases. Innovation in financial services tends to increase their power to create credit, to increase the spread between policy and market interest rates, and therefore to escape from monetary policy control. Charles Kindleberger would point out that bubbles have happened in every known monetary regime, very much including gold, and I think we can summarise the point by saying that financial innovation is the microeconomic reality of the macroeconomic concept of endogenous money.
Why care? The argument here is capable of general application to all kinds of asset bubbles, but housing plays a special role in the economy. It is BULLish – Big, Universal, Leveraged, and Life-Essential. Everyone has to live somewhere and it’s not optional. Because houses are themselves a large investment of capital, building or buying them usually requires credit and lots of it. The combination of the two means that it is big. Therefore, this is a highly leveraged market that touches whole nations in the pocket, unlike (say) technology startup shares or even commercial real estate. Edward Leamer made this point to the Kansas City Fed back in 2007 in a paper called Housing Is The Business Cycle.
How to look at this? The cybernetic tradition, I think, is the right way. Cybernetics, the study of control systems in general, was concerned from the word “go” with the problem of what happens if there are more questions than there are answers. One version of this was imported from psychiatry, the notion of the double bind. A patient is forced by their situation to respond to two mutually incompatible expectations, so that whatever they do is wrong. The result is that they go mad (to be brief), and the shrinks of the time did horrible experiments in inconsistent conditioning with dogs to prove the point.
Various cyberneticians, especially Stafford Beer and Ross Ashby (himself a psychiatrist and quite the pre-ethical review board creep), identified an important principle here: the principle of requisite variety. To exercise control over something, you need a range of responses – a degree of variety – that matches the variety of its outputs.
If its outputs can change along more than one axis, you need at least as many responses. If you want to determine both the air speed and the vertical speed of an aeroplane, you need both the elevators and the throttle. If you want to determine both its course and its attitude, you need both the ailerons and the rudder. If I need to please my mother and my husband…you may see the point. To some extent, you can get away with less variety in the more forgiving bits of the flight envelope. In that case the variety adds to redundancy, which is good. But the problems arise when things become more challenging.
So, we’ve already done too many of them words. The point is, I think, that you can’t have effective monetary policy at the macro-level if monetary policy has to do micro-level interventions. You may not even be able to do those interventions at all, and I have my doubts about doing the whole of macro policy with monetary policy. And we already assume that macro-level monetary policy exists in order to avoid doing micro-level interventions. Central banks getting interested in regulatory intervention do so, in part, to get enough requisite variety so that monetary policy can work.
And if you don’t like this for libertarian reasons, well, it is what it is. We tried that.