What surveys say about inflation

  • The Fed maintains that high US inflation is transitory
  • UK and EA inflation is low, but is it measured correctly?
  • Business and consumer surveys show higher EA inflationary pressures

In an article in The Times on Monday 27th July, my former colleague Simon Ward is quoted arguing that current British inflation is rather higher than the official 2.5% (year to June) figure. Simon essentially says that the reason the inflation numbers are misleading is because they do not take into account the changed spending pattern during the pandemic. (I made a somewhat similar point in a Comment 30th November last year. But Simon makes it better.)

I don’t know if Simon has looked at other economies in the same way, but if he had, I dare say that the results would be similar. 

At the moment there are two key questions with regards to inflation. The first is whether current US inflation is transitory or not. The Federal Reserve continues to argue that it is, markets seem increasingly sceptical. The second question is, why is there no inflation uptick in the other advanced economies.

If Simon is correct, part of the reason for the seeming lack of inflation in, e.g., the UK or the Euro Area (EA) may be that we are measuring the wrong things. As for whether current inflation trends are transitory or not, one of the terms central bankers love to use to show that their policies are correct, is ‘Firmly anchored inflation expectations’.

I have always been somewhat sceptical of this attitude. For instance, Japanese inflation expectations remained firmly anchored even as the Japanese economy slid deeper and deeper into deflation in the 1990s and 2000s. And, in any case, if you have an inflation-targeting and credible central bank, then expectations must be that it will achieve its inflation target over the policy-relevant horizon, because otherwise it would change its policy in order to reach the target. 

Rather than ask in general about inflation expectations (or, indeed, measure them out of what index-linked bonds do), I feel much better information can be found by asking businesses and households what they intend to do in terms of purchases or price changes and what they themselves are experiencing.

For brevity’s sake, I will only look at the United States (the ISM survey) and the Euro Area (EU Commission business and consumer survey). Before looking at the data, it is important to bear in mind that the two surveys show different things. The ISM survey shows prices actually paid for inputs. The index is a diffusion index, calculated by adding the percent of responses indicating they paid more for inputs plus one-half of those responding that they paid the same and then seasonally adjusted. The EU business survey shows what manufacturers intend to do with their own selling-prices over the next three months or so. This nets answers of one type (e.g., raise prices) against the other (cut prices).

Now, what do the indices show?

The ISM manufacturing prices paid index has an all-time (since 1948) average reading of 62.1, Since 2008, the index has averaged 58.3. But over the twelve months from July 2020 to June 2021, the average is 75.6. The index has been over 80 since January this year and in June it rose to 92.1. By way of comparison, the index was last over 80 (for four months) in 2011. In that year, headline inflation was over 3% for ten consecutive months. Similarly, in 2004, a short period of readings above 80 was associated with inflation above 3% later in the year. As for over 90, the index has not been there since 1979, when inflation was still in the double digits. 

The non-manufacturing ISM prices index is not quite as high. Still, the average over the past 12 months is 68.5, compared with an all-time average of 58.8 and a post-2008 average of 58.6. The index began 2021 at 64.2 but has since risen to over 70 and briefly above 80 in May.

Direct comparisons with the past are not perfect. However, the rising trend of the two ISM prices paid indices argue that US inflation is unlikely to be as transitory as the Fed hopes. Yes, the headline rate should fall in late summer/early autumn, but this will mainly be due to a base effect and inflation is then likely to rise again. For how long will it then remain uncomfortably high? US broad money growth has slowed to below 10%; but this is still high by the standards of the past ten years, and there are signs that credit growth is accelerating, which should boost broad money growth again. The likelihood is that US inflation by the end of the year will be at least as high as it currently is, a rate that is already making markets uncomfortable.

What, then, about the EA? The EU Commission survey actually asks all the sectors covered (manufacturing, services, retail, construction and consumer) about price expectations. Historically, the most accurate one is the manufacturing sector question. For the EA as a whole, selling-price expectations have a long-term (since 1985) average of +7 (i.e., per cent of respondents intending to raise prices less those intending to cut them). Since 2008, the average reading is +4 and for six of the twelve months in 2020, the reading was negative. 

That has now changed: the index registered 0 in November 2020, +10 by February 2021, +24 by April and an all-time high of +36 in June before edging down to +35 in July. Historically, a balance of between +15 and +20 has been consistent with an inflation rate of 2% or above. We are now well above that level . There is a similar trend in consumer inflation expectations, which are admittedly not at all-time highs, but the highest in ten years and, again at levels historically consistent with inflation above 2%.

It remains to be seen whether EA inflation will rise much above its current rate of 2.2%. But, based on the business survey data, a rise is more likely than not – and if it does, the ECB has already made clear that it will do little to counter it. Specifically, President Lagarde has said that the ECB will maintain its current policy, not only until inflation reaches its target but also shows that it will remain there for the policy-relevant horizon – even if this means above 2% inflation for some time.

This leaves two final questions:

Why do central banks maintain their easy policies, notably the Fed this week, for all that it hinted of tapering its purchases? One reason may be that the risk/reward of making the wrong decision are skewed. If, as a central bank, you underestimate inflation and it takes off, you feel that you know how to bring it down. By contrast, if you tighten prematurely, you will be criticised both for potentially aborting a recovery and for overreacting against an inflation threat which (thanks partly to your action) may never materialise.          

Second, if business is this concerned about inflation, why do markets remain buoyant? A number of reasons, but among them, first, a feeling that ‘all news is good news’. That’s nice while it lasts, but it can easily turn into its opposite, ‘all news is bad news’. Second and probably more importantly, as long as central banks hold to the view that inflation is not a problem or is merely transitory, they are not going to do anything to tighten monetary policy and stop all that nice liquidity from flowing to markets.