On the firmness of statistical conclusions

I go back and forth on whether it’s worth thinking about macroeconomics at all. The last few decades have been an object lesson in blithely closing your eyes and singing “la la la” whenever anything bad happened. If you were invested in a diversified basket of equities you have done very well. One of the cliches about investing is that we “climb a wall of worry” on our journey towards attractive returns. There have been many moments to worry about, but the long-term result has been very good.

This has led people to invest in index funds to a colossal extent. Index funds track a category of assets automatically; they just buy everything. They are also cheap. Why pay clever people to manage your money when everything goes up anyway? If experts can’t pick winners and there’s no financial apocalypse to protect from, you might as well plump for low fees.

There are a lot of smart people who like thinking and talking about the world, but no one owes them a living. The contrary idea that smart people aren’t necessarily good at investing is an old one: Newton famously lost money in the South Sea Bubble. Today many smart people in asset management, it seems, are better at getting you to believe their ideas than delivering a cost-effective return. So goes the case for investing exclusively in stock indices.

Something niggles at me though. A lot is made of the empirical evidence of market returns. Over almost any conceivable timeframe it can be shown that equities have at least outperformed inflation. They are volatile, but statistically you have been rewarded for that volatility over the history of financial markets. But therein lies the rub.

Financial artefacts like debt, interest rates and inflation are surprisingly old, going back to ancient Mesopotamia. But usable datasets to draw trends from are extremely patchy and the modern series of equity market data is short in comparison to the long sweep of history. One of the longest datasets on inflation that exists is a series of cuneiform tablets citing the price relationship between barley and silver in ancient Mesopotamia. But even this data pales in comparison to the amount of data typically produced by a single run of the CERN particle accelerator. This is all quite apart from the question of whether ‘interest rates’ in the ancient world are really the same thing that the Federal Reserve is tasked with managing, if all the institutional underpinnings are different.

But why compare Sumerian inflation with a particle accelerator? Because the worlds of finance and banking were colonised by physicists in the early 1980s. As all colonists do, they brought their wares and their ideas with them. As a result, financial and business practitioners use the same statistical methods to try and draw conclusions and project the future.  One very common statistical method, the Monte Carlo simulation, is a method for creating a projection about the likelihood of uncertain events. It is credited to Stanislaw Ulam, a physicist working on the Manhattan Project. The team needed a way of predicting the diffusion of loose neutrons through the core of a nuclear weapon as it exploded. As a method it allows us to make predictions about events that are fundamentally uncertain or too complicated to compute deterministically. It is a method now used very widely to project the spread of returns likely for a certain investment. But if the data set created by a day of running CERN is many orders of magnitude larger than the whole dataset represented by the entire performance history of developed market equities, the firmness of the conclusions about asset class returns seems questionable.

There is some insight to be gained in thinking about the kind of game we are playing when, as individuals, we make decisions about where to invest our money. While finance academics and economists might try to discover reliable laws of nature that apply, as individual investors (and those who help them) we are perhaps better off using some simple heuristics. After all, it is not the aggregate experience but our own circumstances which matters most to us. Keeping a steady focus on building wealth over time, making the most of the tax system, ensuring our lifestyles don’t push expenditure out of control and worrying first about the negative impact of inflation are just a few of the principles that, while they may seem homespun and simplistic, are followed by relatively few people. A financial planner is sometimes a professional worrier to ensure you are following these precepts.

If the idea of being helped towards better financial health is an appealing one you can get in touch here.   

Enjoy the holidays.