Inflation Monitoring Comments

I was asked about the commentary in a report "Inflation, Do You Feel it Yet?" by the Interprime Newsletter. I think it follows standard market research conventions, which is aimed to be fairly sensitive to changes in direction. The problem with this sensitivity is that underlying inflation in recent decades in the developed world is quite sluggish. However, as seen in the chart above, standard economic time series have been mangled by the economic disruptions associated with the pandemic.

I will let the reader go through that newsletter. The focus is mainly on various goods and commodity prices, many of which have jumped in the past couple of years.

The advantage of looking at commodity prices is that they are available in real time (or close to real time), and there are no statistical adjustment issues to worry about. The problem with them is that they are now highly sensitive to the global cycle, in particular China. Although we cannot ignore the global cycle, the Chinese economy has been quite perky over the past couple of decades, yet that has not spilled into core inflation in the services-dominated developed economies.

Another issue with some goods prices is that they are tied to investment activity -- in particular, lumber (North American housing) and computer chips (cryptocurrency mining). We know that animal spirits have bounced back in speculative markets -- but we have had ongoing bubbles for a long time without there being much feed through into wages.

It is clear that it is probably a mistake to keep explaining away rising prices as being due to special factors, since supply chain disruptions always show up in particular areas. There is no doubt that there has been at least a one-time upward shock to a variety of prices due to various disruptions, the only question is whether price hikes can be sustained.

In order for goods prices to enter a sustainable uptrend, wages need to also rise to keep the volume of demand stable. The lack of follow-through into wages in the past decades explains why previous price jumps (such as oil price spikes) have not resulted in sustained inflation. Unfortunately, standard wage data has been disrupted by a change in mix (figure at the top of the article). Lower-wage employment was shut down more as a result of the lockdowns, and so there was a mix change that raised wages. (There are various data sources that were supposed to allow us to control for that mix change; I have not dug into them on the basis that I am not attempting to be a forecaster.) At present, there are plenty of anecdotes about labour shortages. At the same time, I dealt with economists who were pounding the table about similar anecdotes in 2010. Various programme changes as well as an unwillingness to take now-risky jobs for a low wage have tightened the labour market, but vaccinations have kicked in the United States (and with a lag, Canada), and the fiscal programmes are rolling off.

React to Markets or the Economy?

Turning back to the Interprime newsletter, one is faced with an interesting problem: what matters for inflation -- actual inflation, or what market participants think? If you are trading inflation breakevens, except for very short maturity instruments, most of your profit and loss is driven by the changes in the breakeven rate, which can be viewed as market expectations for inflation. (Some commentators object to that phrasing, but I think they are on very weak ground. At the minimum, if we use the mathematical definition of expectation, then breakeven inflation are definitely the expected value -- albeit an approximation.)

We are back to Keynes' beauty contest. Market participants react to fast-moving commodity data, and so it is somewhat suicidal to fight that trend. The problem is that if realised inflation does not track commodities, you need to be ready for revisions of expectations when commodity prices turn around. As such, one needs to keep in mind that there are very different audiences for inflation analysis.

You've Been Cancelled!

Finally, my Twitter timeline has been filled with moping by senior "centrist" mainstream economists (Larry Summers being one of them) about being "cancelled" for criticising Biden's fiscal package and/or worrying about inflation.

First of all, it is fairly clear that Democrat-leaning economists are closing ranks around the President. That is not exactly surprising behaviour in the current environment. To a certain extent, some dinosaurs in the Democratic party think it is still the 1990s. American politics is hopelessly partisan, so pretending to be a technocratic centrist is not going to work.

That said, it is clear that Larry Summers is being dunked on because his analysis was downright terrible. He swung from "oh noes, secular stagnation" to his now famous 33%/33%/33% probability distribution. The analysis was undisciplined, and so the response was predictable.

(c) Brian Romanchuk 2021