Pareto again

This post on substack is about Pareto’s Law. This has nothing to do with Pareto Efficiency, the idea that equilibrium should be when any reallocation of productive resources will make at least one actor worse off. Pareto’s Law is about the fact that income and wealth distributions tend to follow a power law, and that specifically, 80% of assets are owned by 20% of people. It’s also about the fact that 90% of returns (for example) come from 10% of investments. The problem, as the post makes clear, is that you don’t know a priori which bucket trades will fall into.

Setting rules, ideally automatic ones, to close losing positions will help. But I don’t really think this is a complete solution. You can be right, but early. Cynics say that this combination is just the same as “wrong” but I’m not so sure. In the long run we are dead, but the market can wake up to new insights in a timescale over which most of us would hope to survive.

Wrap

Oil finally bounced. It was up $2.42 a barrel, nearly 3%, but this takes it only to $84.35, below where it was a fortnight ago. Gold was up too, but only 31 bps. Other commodities were flat. The equity markets were dragged down by $TSLA, down 16% over the last two sessions. This is dramatic, but not systematic, but it did drag some linked stocks with it, notably $ARKK, which was down 2.35%. Bonds were up, presumably in reaction to weakening of equities. PPI was up 0.6% in Oct., an annualized rate of 7.2%, but this didn’t seem to phase the masters of the universe. Nor did tapering of bond purchases. I always think of the discount curve in terms of a real return plus a compensation for inflation. It is also a predictor of future interest rates. It is conceivable that we’re set for ten years of zero or negative (short) interest rate policies, which will fix ten year yields. The demand side will be fixed by QE. This has happened in Japan. It is easy to see why it is preferable to the alternative, which is government default.

Image

Forecasting

Conventional economics is very bad at forecasting, even a short distance into the future. The UK Budget/Autumn Spending Statement is written on the assumption of infallibility of the OBR. Well, a lot of faith is put in OBR forecasts for predicting the future path of the public sector borrowing requirement. But productivity, the most important metric of all, is beyond the capabilities of the OBR to predict. The naive assumption that the UK simply cannot deliver any productivity gains, ever, would be much more accurate than the complex models run by the OBR.

I have great sympathy with the modellers. Complex systems are very hard to model. They tend to be non-linear, and very unstable with respect to initial conditions, especially when so many policies are discretionary and based on whim. We cannot measure the output gap, or global mean temperature, with any confidence. But we can measure unemployment, and CO2 levels in the atmosphere. We should formulate automatic negative feedback mechanisms, which would at least reduce the chances of the economy, or the atmosphere, going off on some sort of runaway and unstable path, rather than leaving decisions to politicians and central bankers. Oh well, I can only dream..

I don’t claim to be an expert on MMT, but the Warren Mosler sect seems to believe in running the economy hot enough to drive involuntary unemployment to effectively zero. I think this involves actually providing work, with a decent wage, to anyone who cannot find a job in the private sector. This would remove the need for minimum wage laws, because it would set a floor under pay, just as central banks set a floor under short term interest rates by paying interest on excess reserves.

I’m nervous about these ideas, because if we allow institutional destruction of price discovery in markets for capital and labour we’ve come close to stabbing a dagger into the heart of the free market, with its magical powers of allocation. But nobody seems bothered by CB’s destroying price discovery for debt. Maybe we should do the same for labour.