The basic drivers of inflation

Inflation has been in an unequivocal downtrend ever since the early 1980s. It would only be fair to open this discussion with a simple why? At the most basic level, the price of goods (or services) can only change if supply of, and/or demand for, the goods in question change. The two curves in chart 2, where D represents demand and S supply, will have to change, and they have indeed changed quite significantly since the early 1980s.

If only life was that simple, but D and S in chart 2 reflect only what I would call the ultimate causes of inflation. An entire series of proximate causes also exists. They are those incidents that cause demand or supply to change. Excess money printing, Reagan’s (and Thatcher’s) war against labour unions and the resulting end of cost-of-living adjusted wage increases back in the 1980s, various oil price shocks, and a change in animal spirits as times got better and better throughout the 1980s and 1990s are only a few examples of proximate drivers of inflation that have changed either supply or demand (or both) in the last 30-40 years.

Talking about excess money printing, I have said it before and I will say it again. QE is not akin to excess money printing. Central banks have two main assets on their balance sheet – cash (coins and notes) and free reserves. QE only affects free reserves, and only an increase in cash would cause inflation to rise. If you still don’t believe me, I suggest you explain why the correlation between inflation (CPI) and free reserves (M2 is a good proxy for free reserves) has been negative throughout this QE saga (chart 3).

Why does inflation continue to confound expectations?

The vast majority of commentators and analysts have failed to predict what would actually happen to inflation. Ongoing QE combined with substantial (and growing) fiscal deficits have led most to predict that inflation (and interest rates) would rise. Meanwhile, exactly the opposite has happened. Why is that?

As always, the reasons are somewhat complex, but here are at least a couple of reasons why inflation continues to fall. The most important one is probably the ongoing productivity shocks led by the digital revolution. When I constructed chart 2 above, I deliberately made the supply side change more substantially than the demand side, as I believe that, since the early 1980s, the supply side has indeed changed more than the demand side has, and the digital revolution is largely responsible for that.

Another reason is what I would call adverse incentives. China, Japan and Europe have all kept industries alive that are not at all competitive, leading to excess capacity around the world and thus falling prices.

A simple example illustrates how productivity shocks may impact the overall economy:

Suppose a whole series of productivity shocks cut in half the cost of making all goods, and assume that everything else remains equal. GDP then falls in half, as GDP = output x price. As a consequence, workers’ income also fall in half.

There is even an economic term to describe such a phenomenon. It is called homogeneous deflation. When deflation is homogeneous, living standards don’t change. Some people have deemed it ‘good deflation’.

What to expect from inflation going forward

Talking about the digital revolution, there are very good reasons to believe that, going forward, automation will intensify, which (other things being equal) should push the supply function further out and suppress inflation.

At the same time, demand per capita is likely to shrink more rapidly in the years to come, as the populace ages. I have, in prior papers, demonstrated a very strong link between demographics and aggregate demand, as older consumers spend considerably less than younger consumers do.

With both the D and the S-functions likely to move in the direction of less inflation rather than more, I can only reach one conclusion, and that is for overall inflation to remain subdued for a long time to come. That doesn’t mean that you can’t experience the occasional ‘pop’, and that you cannot have regional differences. For example, total demand in Europe and Japan is likely to be much more impacted by adverse demographics than is the case in the US, but inflation will almost certainly remain modest everywhere.

One final point worth mentioning. As living standards continue to rise in most DM countries, an ever larger proportion of discretionary consumer spending goes towards services rather than goods, and that is important for one very good reason. Services are on average much more price elastic than goods, serving to flatten the D-curve in chart 2. More money spent on services is therefore likely to suppress the overall rate of inflation further over time.

Investment implications

The obvious one first. As a consequence of very low (expected) inflation pretty much everywhere (with the U.S. being the most likely developed country to deviate, at least modestly), I would not expect bond yields to move significantly higher anytime soon.

Secondly, as far as the alternative lending space is concerned (which this paper is written for), the demand for such strategies is likely to rise, as more and more investors look for alternative sources of income/yield. That is likely to supress expected returns further over time, and could (worst case) drive those returns down to levels that would deem alternative lending relatively unattractive, but we are not there yet.

The biggest risk in the short to medium term, I believe, is that solid demand for higher income could attract more opportunistic people to the space, leading to a deterioration in the overall quality of investment managers, just as it happened in 2007-08.

Niels Jensen

30 March 2016‍

Thank you to Woody Brock for providing the intellectual stimulus for this paper.