Recent Posts

Sunday, August 2, 2015

Comments On Contemporary Inflation Control

Since the 1970s, conventional thinking about macroeconomics was consumed with the challenges of inflation control. This led to entrenched dogma surrounding inflation targeting and central bank independence. However, since the end of the financial crisis, there has been more discussion of the issues posed by slow growth -- although I believe that inflation control would come to the forefront if inflation rates started to rise again. In this article, I give a summary of why I think inflation will probably continue to remain relatively low in the developed countries, even taking into account what I see as politically feasible policy shifts.

How We Got Here

Chart: U.S. Core PCE Inflation Rate (

The chart above shows the "core" (ex-food and energy) rate of inflation in the United States, measured using the Personal Consumption Expenditure (PCE) deflator. (If we include energy prices, the trend is obscured by oil price spikes.) As is well known, the rate of inflation started to creep up in the 1960s, and then hit relatively high levels in the 1970s (roughly coinciding with oil spikes). Note that many developing economies routinely run higher rates of inflation, even the 10% peak would not be considered very high. Other developed economies had a similar experience to the United States, although the magnitude of the swing in inflation varied.

Since the early 1990s, the core PCE inflation rate fell below 2.5% per year, and has stuck there, despite the cyclical swings in the economy. My view is that structural changes in the labour market have caused the disinflation since the early 1990s. This is based on the analysis found within the book Full Employment Abandoned: Shifting Sands and Policy Failures by William Mitchell and Joan Muysken. (I recently reviewed Bill Mitchell's book on the eurozone here.) Unfortunately, it has a ridiculous price, like other academic textbooks.

More generally, the policy mix before the 1990s were not consistent with inflation control. Indexation of payments creates pro-cyclical positive feedback loop. Moreover, the emphasis was on military and infrastructure Keynesianism. Policies created a handful of high-paying "high skill" jobs in infrastructure and military suppliers in response to a unemployment amongst "low skill" workers. The pressure in the "high skill" occupations created inflation, while at the same time unemployment persisted. Minsky discussed this dynamic in the 1960s. (It is not particularly surprising that the big idea seized upon by Democratic economists to deal with "secular stagnation" is a round of infrastructure investment; they are just reverting to historical form.)

The conventional view is that the disinflation is the result of virtuous central bankers pursuing anti-inflationary policies. These arguments are buttressed by models in which is assumed that central bank policy determines the inflation rate. I will leave it to the reader to decide how persuasive such self-referential arguments are. In any event, with policy rates having converged near zero pretty much throughout the developed world, the scope for conventional monetary policy is constrained going forward.

What Should Be Done?

 In the comments section of an earlier article, John asked:
In this regard, I guess what I'm asking is the following. Is there any way of avoiding using the blunt instrument that is monetary policy, with or without a JG [Job Guarantee] (although having a JG probably makes monetary policy less brutal)?
Is it even theoretically possible to put aside monetary policy and solely use fiscal policy (although the delineation of monetary and fiscal policy is problematic), or is it by definition impossible, living in the world that we do? There are few choices and having a blunt instrument is better than having no instrument at all.
I believe that fiscal policy and structural factors within the economy are the dominant drivers of inflation. Monetary policy has at best an indirect effect. Relative interest rates influence currency values (mainly because currency traders seem to like trend-following positive carry trades), but the feedthrough to domestic inflation from currency changes is generally slight. Additionally, monetary policy can presumably cause a recession, aborting the inflationary pressures that build during a business cycle. But at the same time, modern business cycles end because of financial crises that shut down fixed investment; it is unclear whether rate hikes are needed to provoke such a crisis. (It appears that there is a market consensus that rate hikes are needed to bring financial bubbles to an end. This explains the exaggerated attention to the timing of the first rate hike amongst equity investors. But I doubt that bubbles can be sustained indefinitely, even if the central bank remains "on hold.")

Economic policies that are not easily characterised as "monetary" or "fiscal" matter for economic growth and inflation. This means we cannot just look at monetary and fiscal policies to make inflation forecasts.
  • The decline of unionisation is plausibly related to lower wage inflation. To what extent this is the result of government policy, it cannot be traced to fiscal or monetary policy stances.
  • Indexation of prices and wages to inflation will have turn one-time price level shocks (for example, from an oil price spike) into ongoing inflation. The decline of strong unions, as well as the reduced role of the government in developed economies presumably reduced the amount of indexation within the economy over recent decades.
  • Trade policy has led to lowered prices of traded manufactured goods relative to domestically produced services within the developed economies. The economic effect occurred almost solely within the private sector, the involvement of the government was solely regulatory change.
  • In the United States, there has been an ongoing explosion in the cost of going to university, which can only be described as a bubble. Rising education costs has been one of the factors that offset the downward bias to good prices created by international trade. The proximate cause was an "arms race" in spending driven by overreaching administrators who are emulating the empire-building tactics of private sector CEOs. This bubble was partially funded by the rise in university endowments, a side effect of rising asset prices. However, the major source of funding has been from student loans. The government is involved with such loans, but it does not appear to be conventional fiscal policy, as the actual fund flows are from the students themselves. (The government is taking on off balance sheet contingent liabilities by guaranteeing loans.) 
  • The Canadian housing bubble has been driven by administrative decisions taken by the Canadian Mortgage and Housing Corporation (CMHC) - at the behest of the Federal Government. There was a consensus across political parties that Canada needed to make CMHC policies more friendly towards the real estate sector, which entailed a significant relaxation of lending standards for mortgages. The bubble was thus unleashed. The Canadian economy could have been put on a more sustainable growth path by having kept more sensible lending policies. A tightening of standards has occurred has been attempted, but the question arises whether this was done too late. Although low interest rates helped the bubble form, it could have been controlled without affecting interest policy.
With regards to analysing fiscal policy, we need to move beyond highly aggregated models, such as those where all economic decisions are effectively taken by a single household. The economic effect of handing $10 million to poor households is very different than giving a $10 million tax break to a multinational that is already piling up cash on its balance sheet. For this reason, we cannot look at summary measures of fiscal policy (such as the fiscal deficit) and conclude much about the effect on the economy (and inflation, in particular). This provides a theoretical justification for the view that "taxes for revenue are obsolete" (see the discussion on this Neil Wilson article). (If taxes were solely for revenue, the mix of taxes does not matter, just the amount of revenue raised. But since different taxes have different effects on the economy, we cannot say much about the stance of fiscal policy by just looking at aggregate revenue.)

The experience of recent decades shows that the developed countries have a policy mix that is consistent with low inflation. This has been achieved by suppressing wage inflation, leaving the economies with persistently high unemployment and underemployment (and in particular, youth unemployment). Until those structural factors change, it is probably safe to say that inflation will cycle around its current low level.
  • Monetary Policy. Policy rates are stuck near the zero bound, and Quantitative Easing is largely pointless (except to the extent it can reduce spreads, such as the Fed purchase of private assets, or ECB purchase of peripheral debt). Monetary policy is going to be irrelevant for as far as the eye can see.
  • Fiscal Policy. The developed economies can undertake mild expansionary fiscal policies (or austerity policies) without there being a measurable change in inflation. Peripheral European countries such as Greece have achieved mild deflation, but doing that required policies that would only make sense to European policymakers. Extremely expansionary fiscal policies could rekindle inflation, but such policies do not appear to be politically viable. Given the hefty underemployment present within the developed economies, the policies required to tighten the job markets would be well outside of the norms that we are accustomed to. In other words, a fiscal programme that would be widely described as "radical" would probably only be about half of the size needed to cause a measurable uptick in inflation.
The Job Guarantee policy advocated by Modern Monetary Theory (MMT) is one of the few policies that I can think of that structurally change the economy while remaining coherent with other policies that favour low inflation.

The belief that a Job Guarantee would not be inflationary would most likely be contested by its opponents. There certainly be a one-time price level shock, as the Job Guarantee would have strong interactions with minimum wage jobs. Some marginal employers that rely on poorly paid workers doing unpleasant jobs may no longer be viable. (This is exactly one of the objectives of the policy.) But once that initial bit of creative destruction subsides, there is no reason to expect wages to be under continued upward pressure -- unless the Job Guarantee wage is continuously raised. The implication being that aggressive indexation of the wage paid would have to be avoided if the objective is to contain inflationary pressures.

Of course, fixing the Job Guarantee wage for all time would be unfair if the objective of the government is to deliver 2% a year inflation (or whatever). The wage could be periodically updated, but it leaves the programme hostage to political developments, such as we have seen with the minimum wage. A better solution appears to have an automatic growth which follows the level of the inflation target (for example, 2% per year). This way the wage level should remain relevant, but at the same time not act to perpetuate price level spikes.

(c) Brian Romanchuk 2015


  1. Brian,

    I'm bookmarking this post in my "classics" list!

    I would add a lot but will stick to one comment on your take on the JG, which I gather is the main point of this excellent commentary.

    I absolutely agree with you that a JG need not be inflationary. Indeed, its low inflationary properties were considered an advantage in the 1970s by Baily and Solow in their analysis of the macroeconomics of the JG (or public service employment, as it was called then) The problem, in practice, is that mist jobs offered under the program were actually fairly high skilled, which meant they took workers away from private employment, and contributed to the inflation problem at the time.

    The macroeconomics of a JG are sound. Its practical application and administration are the challenging aspects. You state it nicely above.

    However, just a quibble, I'm not sure the extent to which fiscal policy drives recent (say, post-1990) inflation. I think its role has been too small to matter much. I'd limit it to the state of employment and wages.

    1. Thanks!

      With regards to your point about fiscal policy determining inflation, I would characterize it as determining the general range of inflation, but not the exact level. For example, I would characterize fiscal policy as being stable and relatively tight, and keeping inflation closer to 2% than 10%. But once we know that it is around 2%, we need to look at the other factors as to why it is 1% instead of 3% (for example). And since places like Canada and the United States have had inflation stuck in that tight range since the early 1990s, fiscal policy does not tell us much about the wiggles over that period.

  2. Just wanted to point out a potential problem in your post 1970's analysis. Here is the average US inflation by decade since the Fed was created in 1913:

    It looks to me like inflation is always under 3% on average once you control for crazy outliers like the depression, WWII, the oil price shock. Or to put it another way, inflation is not and has not been a real problem for 100 years, so why all the concern? Where is the evidence to support the belief that inflation is this dastardly problem that is always on the verge of getting out of control? One would think that 100 years of data would be enough to satisfy everyone, but as we know, ideology is immune to facts (unless your ideology is like mine where facts and reality are sacrosanct aka science:)

    1. I would first point out that I am trying to be neutral with regards to the desirability of low inflation. I would not be particularly concerned with inflation around 4%-5%, but I imagine that some of my readers in finance would be horrified by the possibility. However, I believe that a "high inflation" policy would be political suicide, and so "inflation control" has to be part of a realistic political programme. (Some people would argue that you could attempt to move the window of "acceptable" inflation, but that would be a huge effort for negligible benefit; you want to reform the real side of the economy, not the price level.)

      As to why others are obsessed with inflation control, I think it is a combination of factors:

      (a) Central bankers are conservative (in some sense of the word), and inflation-phobic by nature. Central banks are also the major sponsors of macro research.

      (b) The political right swept into power based on the "radicalisation of the middle class" (in the words of George Will) that was created by inflation. It is a natural area of strength for those parties, and so they will always rally around the anti-inflation flag.

      (c) The rise of Real Business Cycle models was based on their alleged superiority in dealing with the stagflation of the 1970s. Since those models have no other empirical successes, they have to keep going back to the stagflation story.

      I would agree that it has not been very hard to keep inflation below 3%. In my view, the central banks are taking credit for something that would have happened anyway. I think the reason why they would not embrace a 4% inflation target is that they would probably not be able to hit it. (This is because of nominal rigidity near 0% inflation; I gave a simple model here: link to "Using Nominal Rigidity To Improve An Inflation Model."

  3. Remember, banning (useless) things (e.g. luxury goods) and cutting bank lending can be used to free up real resources.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.