« Leviathan, Exit, Voice and Ontario Teachers | Main | What should a central bank do when producer and consumer prices diverge? »


Feed You can follow this conversation by subscribing to the comment feed for this post.


The other thing that sticky inflation does - it breaks down the super-neutrality of money in the Fisher relation. The Gibson paradox is back - Sargent has been documenting this for the past couple of years.

Now, deficient AD theories have to pick and choose between two competing hypotheses - one is that the neutral rate is prone to violent Keynesian things, and is on its way down secularly, on account of tech stagnation/ monetary policy numbness/ whatever.

The other theory is that the IS curve does slope upwards, even with real rates. Or, monetary policy is about controlling the natural rate itself.

The situation of the great depression, with its deflation that pushed real rates up and thus allowed *Keynesian* and *Monetarist* views of the world to be reconciled and still both be right is no longer there.

Now it's this or that.

In 1982 inflation did fall a lot because the CB said it wanted it to fall. In 2008 inflation didn´t fall because the CB “said” it wanted to keep it at 2%. Inflation is sticky because CB “credibility” has made it so. In that case you can have an AD recession and inflation will remain “sticky” (on target), only RGDP falls.
Now if the CB credibly targets NGDP-LT, NGDP will become “sticky”. A supply shock will temporarily reduce RGDP, and inflation will temporarily go up. But there will be no AD recession because AD will be “sticky”.
In the 60s the focus was on unemployment (4%) – the horizontal axis of the P-C. The result was that the vertical axis went “bazinga”. Lately the focus has been on inflation (2%) – the vertical axis of the P-C. Result: the horizontal axis went “bazinga”.
By focusing on NGDP you stabilize the “whole thing”. And that was the outcome during the Great Moderation.

Nunes comment sounds just like the inflation targeting advocates prior to 2008.

"In that case you can have an AD recession and inflation will remain “sticky” (on target), only RGDP falls."

This is a claim, not an argument. Hitherto if inflation remained steady and you observed a decrease in RGDP, I would have said you identified a AS shock--as what you just described MUST be a shift in SRAS.

What Nick is struggling to reconcile is why there would be unemployment. He would claim: if unemployment rises, you have identified an AD deficiency.

I have not bought entirely into the view the unemployment <=> AD deficiency because there is a short-fall of money and a short-fall of money only. (Though I agree with the weaker claim that a short-fall of money causes unemployment).

We get a hint of this problem from Okun's law. If you all you knew was Okun's empirical data what would you claim is the dependent and independent variable? In Nick's world view, real-output depends on employment. So something which raises unemployment (money shortfall) will also appear to lower RGDP.

The Austrian PSST model I think argues the other way around. Low RGDP growth due to economic coordination failure drives low employment growth which when lower than the rate of population increase means unemployment rises. There we basically have a supply-side issue that is contributing to unemployment.

That´s because for more than 20 years NGDPLT was masquerading as "pure IT"! When Bernanke (an IT proponent) explicitly focused on IT (worse, headline CPI with no consideration of core), things derailed.

Jon: "What Nick is struggling to reconcile is why there would be unemployment. He would claim: if unemployment rises, you have identified an AD deficiency."

Usually, but not always. There can be cases where a real shock causes the natural rate to rise. But this didn't look like one of them.

Ritwik: Yep. "The other theory is that the IS curve does slope upwards, even with real rates."
That's the one I would go for.

marcus: that is my way of thinking about it too. But I wish I had a clearer picture of the SRAS/SRPC curve, and what lies behind it.

Imagine an economy that for 30 years has trend RGDP growth of 3%, but with annual variations in the demand for money of between 0 and 3%.

In universe one: The CB targets keeps base money steady. NGDP varies most years 2% or 3% above and below the 3% trend line for RGDP. Inflation averages 0% but with occasional low inflation or deflation Firms and lines of business that have a flexible pricing and supply model do relatively well in this universe.

In universe 2 the CB target 2% inflation. NGDP grows at 5% every year and inflation is always 2%, There is a relatively smaller benefit to having flexible a pricing and supply model so firms with this attribute make up a smaller share of the economy and firms who can exploit the advantages of steady growth and fixed contracts do relatively well.

In year 31 demand for money increases unexpectedly by 20%. In universe 2 the CB continues to target and hit its inflation target. This is not sufficient to prevent NGDP falling and because of the structure business have adopted this results in firms firing workers and closing factories. In universe 1, because a larger share of the economy has flexible price structures , the fall in NGDP is to some degree matched by falling prices. If the CB steps in with monetary stimulus it may actually be able to keep RGDP on trend without hitting the ZLB.

Your previous post may have been clear, Nick, but it was wrong. Inflation targeting did not fail, because it never promised more than that "price stability....is the best contribution monetary policy can make to economic growth and job creation" ( http://www.ecb.int/mopo/intro/html/index.en.html ). It was the other contributions that failed, or more accurately, never even tried.

I also guessed in the post before that that you were referring to the UK when you mentioned your surprise that sometimes inflation did not fall despite a long recession, and gave an explanation. The UK example also suggests that your present hypothesis is wrong. UK inflation has not been sticky; it has gone up. And this is linked to the long recession because the excessively easy monetary policy that has allowed inflation to rise has also allowed numerous zombie borrowers to survive. Everyone knows that the situation is unsustainable, so they hold back from new commitments. Just like Japan. And less like the US, where house prices at least have been allowed to fall by a salutary amount. I hate to say it, because in general I disagree with their politics, but sometimes the Austrians are right.

It's hard to see how you can ignore the influence of the US. There are, for instance all sorts of opportunities for arbitrage and carry trade in case of divergences.

I wonder is you're doing inflation targeting how is that not sort of like a gold standard? In that it forces the central bank to potentially take actions that are the opposite of what called for? But the idea that by controlling inflation tightly you deprive the system of of a signal that's an interesting one.

Odd thought, say the economy is kind of sucky, but picking up. Potential creditors and creditees think that inflation could well go much higher. That's an encouragement for creditors to lend, anything that will maybe give them a good rate of return in order to preserve their capital. Their counter parties may accept a higher interest rate, betting that higher inflation will make the payments lighter. But if they believe that the central banks is going to slam the brakes on hard at 2.0003%, the whole process is short circuited.


Mentioning the gold standard is interesting. In the 1920's while Benjamin Strong was governor of the NY Fed, it maintained a policy of price stability by buying surplus gold from trade imbalance to prevent inflation (This was before the Fed OMC, which was created by the 1933 act. The NY Fed controlled policy, since it was by far the largest of the banks, and Strong was very influential). This worked, however it thwarted the underlying concept of the gold standard, which was that it was self regulating. What was supposed to happen was inflation in the US would make US goods less competitive. This would rebalance trade, and gold would flow out again.

There was no way to reconcile the two forms of money policy, and a shortage of gold developed outside the US, particularly in the UK, which had been forced by political considerations to restore the gold standard at the pre-WWI price. The high price of gold in UK pounds made for a severely deflationary monetary policy, and eventually the UK was forced to leave the gold standard.

There was another interesting parallel. The US policy restrained consumer prices, but not financial asset and real estate prices, nor did it prevent record levels of consumer debt (surpassed as a percent of GDP only recently). When the Fed became alarmed at the amount of financial speculation (which violated the principle of lending only for "true bills", which was a doctrine supported by Governors like Adolph Miller), its attempts to assert control were ineffective at first, and then arguably aggravated/provoked the crash by raising interest rates, causing liquidity problems for those (many) who were highly leveraged.

The problem of controlling inflation in consumer prices, financial assets and debt at the same time with the customary monetary tool kit is still with us.

Canada may have stable consumer prices, but its consumer debt is the largest as a percent of GDP of all the major developed countries, and it didn't fall dramatically as it did in the US post-2008. Canadian housing prices relative to trend are higher than the record for US prices.

This isn't quite a glorious success for the Bank of Canada.

I rather like Ron Ronson's "evolutionary" story. I think it's logically coherent. I'm not sure it's right though. My hunch is that it would take longer than 20 years for that sort of evolutionary story to play out.

Rebel: Inflation targeting cannot prevent real shocks. But it was supposed to prevent nominal shocks, and prevent real shocks turning into nominal shocks. It was supposed to prevent deficient demand recessions (unless the central bank's crystal ball failed, and it failed to prevent inflation falling below target). Now you could argue that there is no deficiency of aggregate demand in the UK, rather there is excess demand, and that's why inflation increased. But that doesn't look plausible to me. It is (I think) easier than normal to buy stuff (both goods and labour) in the UK, and harder than normal to sell stuff.

Inflation targeting....was supposed to prevent deficient demand recessions.

No it was not, Nick. I gave you a quote from the ECB; you give me a quote that supports your claim that central banks were trying to prevent deficient demand recessions.

How would I tell the difference between a demand-deficient recession and an excessive price recession anyway? As a UK citizen, it seems to me to be plain that our problem is that wages are just too high. Industry is relocating to Poland etc. Now I grant that you can fix this with a bit of unanticipated inflation, but what does that do to moral hazard? The UK already has what I describe as a cumulative, two-sided, cultural moral hazard problem (ie the UK housing market). "Cumulative" because it involves aversion to house price falls whatever the starting level; "two-sided", because it involves a disincentive to stay out - ie low deposit rates - as well as an incentive to get involved - rising asset prices - and "cultural" because it has become folk wisdom rather than a deliberate calculation. We cannot afford to make that worse. If wages are sticky, better to try to unstick them (eg by improving management-worker co-operation as in Germany) than inflate other prices.

Rebel; your ECB quote does not contradict what I said. The main way that IT was supposed to make ".. the best contribution monetary policy can make to economic growth and job creation" was by smoothing out recessions.

If you Google "divine coincidence" you will see papers (e.g. Blanchard) discussing whether or not, or under what conditions, keeping inflation on target would also keep the output gap to zero.

No Nick, the ECB quote is plain. It is PRICE STABILITY, not the process by which price stability is achieved, that makes the contribution. And that is because it drives businesses and hopefully governments to seek real solutions to the challenges they face. I was a central banker in the years leading up to the crisis, so I like to think I know what they were trying to do.

Actually, Blanchard's paper notes that central bankers do not believe in divine coincidence, and basically says that they are right to do so, since those models in which divine coincidence applies involve unrealistic assumptions, so if anything it supports my argument.

From Blanchard (2005):

"The property just described can in turn be traced to the absence of non trivial
real imperfections in the standard NK model. This leads us to introduce one
such real imperfection, namely real wage rigidities. The existence of real wage
rigidities has been pointed to by many authors as a feature needed to account
for a number of labor market facts (see, for example, Hall [2005]).We show that,
once the NK model is extended in this way, the divine coincidence disappears."

And there's always this more general criticism of DSGE results:

Solow (2010) has this to say about the DSGE models:

'They take it for granted that the whole economy can be thought about as if it were a single,
consistent person or dynasty carrying out a rationally designed, long-term plan, occasionally
disturbed by unexpected shocks, but adapting to them in a rational, consistent way....The
protagonists of this idea make a claim to respectability by asserting that it is founded on what
we know about microeconomic behavior, but I think that this claim is generally phony. ...
The DSGE school populates its simplified economy – remember that all economics is about
simplified economies just as biology is about simplified cells – with exactly one single
combination worker-owner-consumer-everything-else who plans ahead carefully and lives
forever. One important consequence of this “representative agent” assumption is that there are
no conflicts of interest, no incompatible expectations, no deceptions.
...the basic story always treats the whole economy as if it were like a person, trying consciously
and rationally to do the best it can on behalf of the representative agent, given its
circumstances. This cannot be an adequate description of a national economy, which is pretty
conspicuously not pursuing a consistent goal.'"

Very creative post nick. I assume you meant inflation was sticky in the pre-lehman 2008? Inflation fell off a cliff in Post-Lehman 2008.

Henry; thanks. If you look at these graphs, you see that total inflation blipped up then blipped down then recovered to roughly normal in 2008, but core inflation just kept on rolling along, as if nothing had happened. (I'm talking Canada, BTW). There was a recession, but inflation didn't fall.

Hmm. You might be onto something here. Based on the graph below (change the dates to 1998-2013), I noticed that inflation used to be more volatile as measured by the monthly % change, but has been rather calm in this recession. An alien might have thought the 2001 recession was more severe! Incredible credibility.

TBH, I always felt that a fixed target, whether inflation or NGDPLT/capita is too arbitrary. A CB is just another portfolio manager. Its job should be to maximize the "Sharpe ratio" of the NGDP (i.e. maximize the NGDP growth per unit of NGDP volatility). The optimal level of the NGDP should vary over time depending on each economy's development phase and underlying characteristics. The CB can gauge the volatility of each sector of the economy, when one is too far below the trend volatility, the CB steps in to purchase financial instrument specific for that sector only (e.g. MBS for housing, sector specific bond ETFs, and sector specific equity ETFs). These are very targeted tools. Note that this is very different from targeting asset prices.

Interested in your thoughts on optimal CB'ing.


I just got through "Inside Job", a rather excellent documentary of the current recession.

Pertinently, it highlighted some very disturbing conflicts of interest by academics, namely publishing economic overviews praising a country's stability (for a fee from the national Chamber of Commerce, undisclosed), consulting arrangements with firms in areas which have benefited from the academic's policy recommendations and positioning, and the reverse influence of said consulting arrangements determining what an academic does and does not study or recommend.

I would recommend, as a Worthwhile Canadian Initiative, that WCI start a disclosure page for blogger's financial relationships (I expect they are dreadfully boring, a pension, government funded professorship, but a private chair would be worthwhile to disclose, for example). Not because I doubt anybody's probity, I don't, but to set a good example.

What matters is to say that you do or do not have specific consulting relationships with industry, or hold major financial positions directly in chartered banks (not in pensions or mutual funds) or hold a chair funded by a chartered bank donation. Or maybe you have a Canada Chair in something.

It would improve your own already high standing, Nick, by being a good example and leader in best practices.

Very good article. I would just dispute one claim (not that important to overall message): "Gasoline prices went up a bit, but they are stripped out of core inflation, and core inflation looks flat too".

Yes, gasoline prices are stripped from core inflation nevertheless gasoline prices will affect other prices given enough time as gasoline (and energy) is important input. You could invent an inflation measure without wages, but it would not prevent changes in wages to spill over into prices of everything else that uses labor as input. So in short, core inflation may be useful short-term metric used for forward guidance, but it does not help with your analysis of decades long trends.

Determinant: "What matters is to say that you do or do not have specific consulting relationships with industry, or hold major financial positions directly in chartered banks (not in pensions or mutual funds) or hold a chair funded by a chartered bank donation."

FWIW: I do own a couple of hundred thousand worth of shares, mostly Canadian shares, and that usually includes a Canadian bank or two. It used to be TD, now it's a couple of tens of thousands worth of BNS. I don't do consulting. I worked for the BoC 10 years ago on my sabbatical. I used to get royalties, and still get a little, from when I was a co-author on the Canadian edition of Mankiw's intro text. I have a pension at Carleton, some farmland and real estate in England, and my house here. Occasionally I get emails wanting me to promote someone's book. A couple of times I got sent a copy of a book. Nobody has ever offered me money to shill for them.

I am one of the most widely read Canadian economics bloggers, and I know some people in Canadian banks read me, but nobody has ever offered me anything, except an invitation to speak with them. OK, I have eaten 3 free lunches plus a glass of wine when I talked to a lunch-group of retired public servants. And coffee and doughnuts when I talked with the Finance minister's people.

Some people seem to be interested in what I say, but nobody seems at all interested in influencing what I say, except by arguing with me, or saying I should do a post on topic X.

Maybe I'm naive, or don't move in the right circles, but that world seems very different from the world I live in.

Canadian GDP and US GDP are highly correlated.

You can't say what Canada would have to have done to successfully target NGDP and decorrelate. At some point of policy divergence, arbitrage and a carry trade would have appeared, and there's the question of the effects on the exchange rate.

OTOH you can say based on the US example that use of unconventional monetary policy wouldn't have been in conflict with their inflation target. They could have supplied liquidity faster and more decisively than the FED did. But if customers in the US cut back on buying Canadian goods, there have to be limits to what the bank can do to compensate. The US economy is much bigger than the Canadian one.

I think you need an example from a different country to test your thesis.

JV: Thanks. "Yes, gasoline prices are stripped from core inflation nevertheless gasoline prices will affect other prices given enough time as gasoline (and energy) is important input."

Whenever we say "X affects the price level", we are always making some implicit assumption about what other things we are holding constant, especially monetary policy. If the monetary policy is to hold core inflation constant, then nothing should affect core inflation.

Peter N: "But if customers in the US cut back on buying Canadian goods, there have to be limits to what the bank can do to compensate."



You asked why I thought "if customers in the US cut back on buying Canadian goods, there have to be limits to what the bank can do to compensate"

Canada exports to the US run around 18% GDP with imports a bit less. We know from recent research that investment falls off at a high multiplier relative to consumption and employment. The banks job would be to prevent this fall off. I'll assume printing money and hiring every unemployed person in Canada would be fiscal policy. This would, of course, have to work because of the way GDP is defined.

A combined fiscal/monetary policy always has to work by definition. Yet we don't see much of this. The reason other than conventional wisdom is fear of side effects - inflation, crowding out, trade balance, exchange rates, asset inflation...

The next option would be QE and massive liquidity support. This would almost certainly been a good idea and would have reduced the GDP dip, but that's the supply side. Given dependence on the US, the demand side is a harder problem. From where do you get the necessary demand (which has to have the right mix, since only the concrete industry buys aggregate).

Somebody has probably already modeled some of this, or they have a model that could. Set a graduate student to research the problem. You're talking about important public policy issues.

An example. Canada exports enormous quantities of wood products to the US. There's no way to stimulate ordinary Canadian demand enough to compensate for a dip in US purchases. Canada could subsidize the industry so it could cut prices to China and steal enough market from other countries' wood products, but somebody would complain to the ITC, and there could be retaliation.

So the government could replace ordinary demand with government spending at the macro level, but many industries wouldn't benefit. This could be compensated for in aggregate by enough spending, but the downside here is pretty clear. This is structural crowding out rather than competitive crowding out, but it's still ugly.

Alternatively, the government could buy the surplus lumber and stockpile it or maybe build every high school in Canada a wooden hockey facility. This would seem to work better, but it gives me an uneasy feeling. You could create an unkillable monster like agricultural price supports. I lack faith in free lunches.


I know you're widely read, but change doesn't happen at the top, not immediately. I think it can only improve your credibility and improve the quality of Canadian economics debates. Often it's the upright who have to shame the others into correcting their errors. You're the upright.

For example, The Motley Fool is a financial website in the US I used to read. It caters to small investors. When the SEC proposed Regulation FD in the 1990's which would ban analyst-only conference calls and hence indirect insider trading by analysts and their clients, analysts were all against it. TMF mounted a campaign for it. Arthur Levitt, the SEC chair was floored to see the flood of personal letters from TMF readers. The SEC is not used to that sort of popular reaction. It was critical in getting the measure passed.

TMF still has a strict disclosure policy about when its writers write about stocks in their holdings. To change the bad you start with the good.

About robots and the economy from the estimable Felix Salmon's blog with links to some interesting commentary:


What can we expect?

Nick, I'm going to give you a possibility or three whic you haven't considered.

Suppose that there's plenty of cheap money "in the economy", but most of it is held by the richest 0.1% of the population. Also, they're the only ones with access to the discount window.

Everyone else is in so much debt that they're afraid to take on any more debt. Bankruptcy is difficult and wages aren't going up.

You would see a demand-deficient recession.

Would prices drop? No, prices are set by the business owners, the rich, the 0.1%. Most business owners are not in commodity markets, most of them are in oligopolistic competition, nearly all of them are price-setters, and there are not enough competitive effects to cause price drops.

Would wages drop? Wages are sticky, it's hard for them to drop.

Please note that in fields where buyers solicit *bids* for work, prices ARE dropping. Deflation is already happening in these areas.

As for actual natural resource commodities, where competition would bring prices down, pretty much all of them are facing supply constraints: you can look this up, but peak oil and global-warming-induced droughts are pretty easy to notice if you're paying attention.

I'm saying: unbundle the aggregates. Look at the *distributional* situation. Look at *market power*. Look at the *individual markets* in each sector. Look at how businesses *actually behave* -- price-setters vs. price-takers -- rather than believing in the voodoo of the supply curve (which does not correspond to real business behavior).

There's a bunch of overlapping phenomena here and you don't need to resort to "expectations" to explain the situation.

Nathanael: you are speaking to the guy who invented macro with imperfect competition. (OK, that's a slight exaggeration, but I did beat Blanchard and Kiyotaki to publication by a whole couple of months.) It doesn't work like you say it does.

RebelEconomist, "price stability" is much more general as a concept than IT. For instance, although NGDPLT is generally considered to be a policy which stabilizes nominal incomes, you could also phrase it as "price stability, modulo supply shocks". George Selgin and Steve Horwitz have discussed this extensively.

If preventing demand-side recessions is _not_ the most important contribution monetary policy can make, then what is it? You refer to giving good incentives on businesses and governments, but why wouldn't NGDPLT also have good incentive properties?

Since you've been talking a lot about interest rates recently, you might like this from Izabella Kaminska at FT Alphaville on the relationship between negative interest rates, commodity futures prices and production. The idea is that with negative interest rate producers do better holding on to commodities rather than selling them. There are other consequences and this may be a way of gauging real interest rates.


The joys of backwardization and contango.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad