I’ve been reading a few books lately on aspects of monetary policy, and might come back to write about some or all of them. But there has been quite a bit of discussion recently – on economics Twitter, and blogs – about a new working paper from a senior Federal Reserve researcher, Jeremy Rudd.
Judd’s paper runs under the title “Why Do We Think That Inflation Expectations Matter for Inflation? (And Should We?)”, which seems like a worthwhile question, especially at the moment when – and especially in the US – debates rages as to how just how transitory (or otherwise) the recent surge in inflation rates will prove. It is common to hear central bankers opining about how much may turn on whether these higher headline rates get into (alter, affect) the expectations about future inflation of firms and households.
If you come at these things from a New Zealand perspective, the most remarkable thing about the paper is probably that it was published at all. How well I recall getting rapped over the knuckles, with severe expressions of disapproval from Alan Bollard, when I used a quiet New Year’s Eve in the office a decade ago to write a short discussion note, circulated in that form only among a dozen or so senior colleagues, in which I had the temerity to suggest that we might consider advancing the case for a legislated Monetary Policy Committee. Not exactly radical stuff, given that it was the way most countries did things (and NZ now does things). When somewhat later word of the paper got out – a Treasury official who had a copy mentioned it in a reference in a paper that was OIAed – the Bank insisted on fighting all the way to the Ombudsman (where the Bank won) to prevent release. Sceptical perspectives on LVR restrictions, before they were put in place, were equally unwelcome, even internally. And when I say “unwelcome”, I don’t mean anything of the sort of “interesting arguments, but I’m not persuaded because of x, y, and z”, but much more of a “back in your box” sort of thing.
And both of those examples are just about internal circulation. I’m pretty sure that no Reserve Bank analyst, economist, researcher or the like has ever published anything that made the hierarchy even slightly uncomfortable in the entire 31.5 year history of the modern (operationally autonomous) Bank. Consistent with that, of course, even though we now have a Monetary Policy Committee with non-executive members, it operates totally under the thumb of the Governor and nothing of a diversity of view is ever heard. The contrast to, say, the Bank of England, Sweden’s Riksbank, or the Federal Reserve is stark.
I don’t want to appear all starry-eyed and naive here. Every institution – every central bank – has its limits, and even the more-open places seem to be quite a bit more open than they were. But it is inconceivable that anything like Rudd’s paper could have been published by the Reserve Bank, even though in many respects it is much less radical than some commentary has tried to suggest, or than the tone Rudd affects on page 1, with his
Economics is replete with ideas that “everyone knows” to be true, but that are actually arrant nonsense.
One natural source of concern is if dubious but widely held ideas serve as the basis for consequential policy decisions.2
I have no idea of Mr Rudd’s politics, but like many readers I was intrigued by the footnote to that sentence
2 I leave aside the deeper concern that the primary role of mainstream economics in our society is to provide an apologetics for a criminally oppressive, unsustainable, and unjust social order.
To be honest, I used to edit Reserve Bank research and analytical papers etc for publications and – keen on openness and diversity as I am (see above) – I’d have insisted that sentence come out. Attention-grabbing but quite unrelated to the substance of the paper (or the functions of the Bank) would no doubt have been the gist of my comment.
But what of the substance of the paper? There isn’t really much there that is new. Quite a bit of it is about the limitations of how formal macroeconomic models capture, and ground, a role for inflation expectations. I don’t think any of that will have surprised most readers, or disconcerted anyone who has been associated with the actual conduct of monetary policy in recent decades. Perhaps you might be slightly disconcerted by his point that the models often seem to put more weight on short-term expectations (where surprises/shocks can generate real consequences) but that “one of the few shreds of empirical evidence that we do have suggests that it is long-run expectations that are more relevant for inflation dynamics”.
But even then I’m not sure that you should be disconcerted, in part because nowhere in the entire paper are interest rates mentioned, or financial instruments, and I (at least) have always thought of the role of inflation expectations as potentially most important in the context of a willingness to borrow (in particular) given the prevalence still of long-term nominal debt contracts (particularly so in countries such as the US where long-term fixed rate debt is a large chunk of the market). A 5 per cent mortgage rate is one thing if I’m working with an implicit, perhaps even unconscious, sense that normal inflation is 5 per cent, and quite another if I’m working with 0 per cent inflation as my norm.
There is a school of thought (class of economic rhetoriticians) who will assert, sometimes quite strongly, that in the long-run inflation expectations are the only determinant of inflation. I had a boss for some years who regularly ran that line. And, to be sure, you can set up a model in which it is true, but that model typically won’t be very enlightening at all, since “inflation expectations” (however conceived or measured, and measurement is a real challenge) don’t occur in a vacuum. If we had the data in the early 1980s, New Zealand inflation expectations might well have been about 12 per cent (say), but inflation expectations were that high because of some mix of (a) the government and the Reserve Bank not having done much to get inflation any lower, and (b) the government and the Reserve Bank not being thought likely to do much in future to get inflation much lower. Policy tended to validate the expectations, but it wasn’t the expectations that determined inflation, but the policy itself. When policy stopped validating those high expectations, they came down (albeit often quite slowly, sensibly enough (on the part of those forming the expectations).
Those misperceptions can matter. When we were trying to get inflation down (to something centred on 1 per cent) in the late 80s and early 90s, no one put much weight on the chances of success. Quite probably many of us didn’t either (I recall a conversation with the-then Westpac chief economist in which I suggested that I’d be reluctant to bet on inflation averaging below 3 per cent for the following 20-30 years). That made it harder (and costlier) to get actual inflation down, but – through some mix of good luck, bureaucratic resolution, and close-run-thing political commitment – we did. And indications of expectations about future inflation followed. A 14 per cent bank bill rate by the mid 1990s no longer meant what it had in 1988, when the inflation targeting scheme was first hatched.
On the other hand, it seems likely (but I’m more open on this) that during the period over the last decade when core inflation was persistently low – repeatedly surprising the Reserve Bank, among others – the fact that indicators of inflation expectations mostly tended to hold up nearer the target midpoint may have helped, a little, avoid more of a fall in inflation itself (although even this is arguable since had inflation expectations fallen away more sharply and obviously, the Reserve Bank might well have used policy more aggressively than it did, including getting unemployment down earlier/further).
One of the other limitations of Rudd’s paper is that there is barely any mention of any country’s experience other than that of the United States. Of course, he is American, writing in an American institution for a primary audience that is America, but…..data. In truth, there just is not that much data in any individual country (because no matter how many series and how high-frequency the data, there are only so many genuine cyclical episodes to study). In almost no other country in the world is it conceivable that someone would write such a paper without looking beyond their own borders, and own central bank. Even for the US, it should be more important, since the Fed focused on an index which doesn’t have a great deal of general public visibility, whereas many other inflation targeters will at least start from the CPI.
For me – as someone with (mostly) a policy focus – the most significant part of Rudd’s paper was the last few pages on “Possible practical implications” and “Possible policy implications”. I had a tick beside this paragraph
Another practical implication is rhetorical. By telling policymakers that expected inflation is the ultimate determinant of inflation’s long-run trend, central-bank economists implicitly provide too much assurance that this claim is settled fact. Advice along these lines also naturally biases policymakers toward being overly concerned with expectations management, or toward concluding that survey- or market-based measures of expected inflation provide useful and reliable policy guideposts. And in some cases, the illusion of control is arguably more likely to cause problems than an actual lack of control.
But for all the glib rhetoric that sometimes comes from senior central bankers, I wonder how many – if any – practical central bankers operate as if they really believe that everything (about future inflation) rests on inflation expectations. I’ve had many criticisms of the Reserve Bank of New Zealand over the years, but not even Don Brash acted and operated policy as if that was his view, and certainly none of his successors have.
Perhaps more interesting was this
Related to this last point, an important policy implication would be that it is far more useful to
ensure that inflation remains off of people’s radar screens than it would be to attempt to “re-anchor” expected inflation at some level that policymakers viewed as being more consistent with
their stated inflation goal. In particular, a policy of engineering a rate of price inflation that is
high relative to recent experience in order to effect an increase in trend inflation would seem to
run the risk of being both dangerous and counterproductive inasmuch as it might increase the
probability that people would start to pay more attention to inflation and—if successful—would
lead to a period where trend inflation once again began to respond to changes in economic
It harks back a bit to the definition of price stability Alan Greenspan once used to give, that it is when inflation isn’t a consideration for people (firms and households) in the ordinary course of their lives, but also seems to be a bit of dig at the current FOMC policy of aiming to run core inflation above target for a time. I’m probably more sympathetic to that approach than Rudd – including for New Zealand after a decade of undershooting the target – but his comment is a perspective that should be taken seriously.
HIs final main point is this
A related issue is more pragmatic. In some ways, the situation that arises from a focus on
long-term inflation expectations is similar to one in which a policymaker seeks to target a single
indicator of full employment—for instance, the natural rate of unemployment. Like the natural
rate, the long-run expectations that are relevant for wage and price determination cannot be directly measured, but instead need to be inferred from empirical models. Hence, using inflation
expectations as a policy instrument or intermediate target has the result of adding a new unobservable to the mix. And, as Orphanides (2004) has persuasively argued, policies that rely too
heavily on unobservables can often end in tears.
People (including central bankers) fool themselves if they think that survey responses, or implied breakevens from inflation-indexed bond markets, “are” inflation expectations (for the economy as a whole) themselves. They are what they are, and always have to be taken with at least some pinches of salt. In New Zealand, for example, household surveys regularly produce numbers suggesting households expect to average between 3 and 5 per cent over periods 1 to 5 years ahead, but no one has ever taken those absolute numbers seriously (there is little or nothing else anywhere in the economy suggesting that whatever people tell surveytakers they act as if they think inflation will be this high). At best, they are indicators, straws in the wind, and sometimes what look like good relationships then no longer do.
As an example of the latter, the Reserve Bank economics department at times articulated a line that the two-year ahead measure of inflation expectations in the Bank’s survey of informed observers) almost was a measure of core inflation itself.
It held up quite well over the best part of 15 years, until it didn’t. It left the Bank too complacent through the following decade (but the error could equally have run the other way).
I guess my bottom line is that one should rarely put too much weight on any specific indicator, and perhaps especially ones that are hard to observe (or to know what one observes actually means). If we see medium-term inflation expectations – among informed observers – at 5 per cent, we (and central bankers) should be disconcerted, but it is highly unlikely that such an inflation expectations number will have been the first sign of trouble.
Changing tack, what of current monetary policy in New Zealand? There is an OCR review tomorrow. Expectations measures here don’t appear troublesome at all – even the inflation breakevens are getting nearer the target midpoint than we’ve known for some years. But core inflation has been rising, unemployment had fallen quite low, and a lot of indicators pointed to emerging capacity pressures. All that was, of course, before the latest Covid outbreak.
I still think there is a very good case for things the Reserve Bank MPC will not do tomorrow: discontinue the Funding for Lending programme, and start a well-signalled programme of bond sales to reverse the LSAP programme. But what of the OCR itself? I won’t be particularly critical of the MPC if they do raise the OCR by 25 basis points tomorrow, but I think if I was in their shoes I wouldn’t. There is a full forecast round and full MPS at the next review and a lot of uncertainty about the Covid outbreak and is its implications (as well as some emerging downside global risks, notably from China). Yes, monetary policy works with a lag, but the starting point for (core) inflation is not so high that we need to be in a hurry to raise the OCR in such an uncertain and unsettled climate. We will know a great deal more – including about vaccinations, and hopefully about exit pathways – on 24 November than we do now. If all is going really well by then, or if core inflation in the CPI later this month is really troubling, there need be no problem with going 50 basis points then, if the data support such a call. But I wouldn’t be rushing right now.