« Tightening money and widening Eurozone spreads | Main | Provincial Government Health Spending: The Equity Dimension »

Comments

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

Nick: "First they came for the 1%..."

I shouldn't have said "silly." Sorry. But the Third Reich? And here I was, just thinking about Scandinavia.

"You are talking to someone who is "the other". We look just like you "regular folk"."

I never assumed you were anything else. Why do you assume I am?

I often think of the great post you wrote about the three countries Robot, Labour and Land (I think). In the end, all goods get split via negotiation between Robot and Land (and all in the hands of Land if they can divide the negotiating power of Robot). Labour (humans) gets nothing. When we actually get there, some humans, of course, will be holding the robots and the land (entirely by inheritance). That strikes me as unfair. Of course, that's far in the future, but I think the process is already under way. Everywhere I look I see arbitrary initial distribution of wealth and seeking of rents. If the end game is perfectly unfair, the current state only differs by a matter of degree.

It seems perfectly reasonable to me to say that because of weak bargaining power, any benefits from inflation accrue to the rich, and similarly because of weak bargaining power, any disbenefits from deflation fall on the poor. You mustn't forget about the power.

And am I the only one here to find shameful the behaviour of someone who declares that they're "an expert on what non-economists think about economics" and deciding that what they believe is a "fallacy", while simultaneously forgetting the existence of "foreign trade"? And another economist agrees - Scott Sumner. This is pure hackery - "The simpletons are committing a fallacy (so long as we assume a closed economy)!".

Economics is in a bad way if some of its main proponents can't recognise very simple concepts like this.

@Nick: "And the usual estimate is that something like 2% inflation is enough to avoid most of the problems this zero lower bound on wage changes could cause."

Thanks, I'm aware of the overall argument about 2% being 'acceptable' (although another justification for this is substitution/new goods/quality improvements, etc, not just the bit of extra distance from inflation).

What I was trying to make explicit is that the 2% is based on a kind of rule-of-thumb based on an assumption of a 'normal' economy, where effectively the normal is defined as 'having some smallish distribution around where prices should be'. Call that distribution whatever you like, but the larger and more random (for goods that are linked to each other in some way, either substitution/elasticity or linked inputs in production) the distribution is, the more likely 'higher' inflation would be needed.

And hence 'high' inflation (compared between countries in absolute terms) is not obviously meaningful. Prices in country X are highly distorted? Perhaps somewhat higher inflation is needed, and monetary policy prescriptions should be adjusted.

@Reason: thanks, agree. There are lots of good reasons for the stickiness.

Edit: whoops, meant to say 'extra distance from deflation.'

A different Alex: "It seems perfectly reasonable to me to say that because of weak bargaining power, any benefits from inflation accrue to the rich, and similarly because of weak bargaining power, any disbenefits from deflation fall on the poor."

Lots of things seem perfectly reasonable to us that turn out to be wrong, when we think about it. If someone has weak bargaining power, why should the value of money have any effect on whether they get a bad deal? Just because they have weak bargaining power, and the other side has strong bargaining power, doesn't mean either side gets confused by changes in the value of money. If they bargained in pennies, rather than in pounds, would it help or hurt them any?

And yes, amazingly, I do understand that an improvement in the terms of trade might make us better off, because we can buy more imports with our exports. But there's also something called the exchange rate we need to consider. If the central bank eases monetary policy, and this causes inflation, how does that affect the exchange rate? Because whether or not it makes us worse off or better of depends, in part, on whether the exchange rate adjusts in proportion. Amazingly, some of us economists actually try to think about and answer questions like that.

And we also recognise that if you want to think clearly about questions like that, you first need to clear your head of conceptual rubbish like the inflation fallacy. So we start out thinking about a closed economy, where it's easiest to see why the inflation fallacy is rubbish, then we move on to the open economy and ask whether it's any different.

And when we have cleared all this conceptual rubbish out the way, we can go on and consider the sort of interesting questions that people like GA want to talk about.

GA: "What I was trying to make explicit is that the 2% is based on a kind of rule-of-thumb based on an assumption of a 'normal' economy, where effectively the normal is defined as 'having some smallish distribution around where prices should be'. Call that distribution whatever you like, but the larger and more random (for goods that are linked to each other in some way, either substitution/elasticity or linked inputs in production) the distribution is, the more likely 'higher' inflation would be needed."

Given the (not unreasonable) assumption that some prices (or wages) have some sort of absolute downward nominal rigidity, I think your argument makes sense.

If someone has weak bargaining power, why should the value of money have any effect on whether they get a bad deal?

Well, of course, if someone has weak bargaining power, they get a raw deal whatever price inflation. But as Alex Harrowell as been pointing out, if your NGDP target assumes that demand increases via wage inflation, and all the vast majority of the financially constrained ever get is price inflation without much increase in their wages, then this could choke off recovery. When it's said that there's a lack of supply of money relative to demand, who do you think it is that has the excess demand for that money? Those with very little, like the unemployed, and those on low incomes who tend to work for others and consequently have low bargaining power. (It also seems reasonable to say that the lower your wage the less bargaining power you have, all else equal).

Now, my response to this would not be to dismiss it out of hand as a fallacy as you have, but question why, when we're implementing an NGDP target, we can't have the CB bypass the banks and the corporate sector more widely, and simply do the helicopter drop - hand money out to the ones who have an excess demand for money, overwhelmingly the poor. Which would increase their nominal income, and help make sure that the benefits of wage inflation is shared more widely, rather than just hitting millions with price inflation with no increase in their wage to go with it.

This leads me on to a second point, which is why do we even want to see the economy grow in the first place? I don't know about you, but the left (at least the non-"dark green" left) wants economic growth so that the vast majority can prosper together. We don't want to see a situation where, either through the CB, fiscal policy, or a combination, the economy recovers, growth returns, but the benefit of that growth only accrues to the top. If, in the process of a policy to restore growth, the vast majority of the population gets shafted, what was the point?

So we start out thinking about a closed economy, where it's easiest to see why the inflation fallacy is rubbish, then we move on to the open economy and ask whether it's any different.

But that's not where this post came from! In a response to a previous post, the other Alex pointed out that in Britain, wage growth is falling behind price inflation. You responded that his idea was that "Inflation is bad because it means our real incomes/real wages are lower" and therefore he was committing the "inflation fallacy". You have since explained what you mean by that, both here in this post, and over in the comments on A Fistful of Euros. At no time did you say, "iff we had a closed economy, then Ron Paul types would be engaging in a fallacy if they said inflation decreased average real wages". Since we're not talking about a closed economy, how can Alex Harrowell have been engaging in a fallacy? Moreover, since there are no closed economies, then there is no such thing as the "inflation fallacy" (to be clear - I'm fine with people assuming a closed economy, then dropping that assumption, but that ain't what you did).

And actually, the fact that Britain is not a closed economy is important when you realise that the price inflation right now (aside from the VAT rise in January) is coming from abroad. It's not a fallacy to point out that the average Briton is getting poorer in real terms whereas foreign energy producers are getting richer.

Do we see such profound disagreement about basic terms in the natural sciences like has happened here? My point is - we do not. That should tell us something about economics.

A Different Alex: "Do we see such profound disagreement about basic terms in the natural sciences like has happened here? My point is - we do not. That should tell us something about economics."

Let me spring to the defense of economics. Well, sort of. ;)

In physics we talk about pressure. We can measure it. But when we get down to the level of individual molecules, we no longer observe pressure. What we observe is random motion, such as Brownian motion.

In economics we talk about inflation, but we cannot measure it. What we observe are price changes of different goods and services that are non-randomly related. We can deal with this non-randomness in various ways, which give us different measures of inflation. One way is to focus on money and ignore everything else. That gives us the idea of inflation as the growth of the money supply. Other ways give us the Consumer Price Index, the Producer Price Index, Headline Inflation, Core Inflation, etc.

When Nick says to make P a dimension in a graph, he is talking about something that is not well defined. That does not mean that doing so is a useless exercise, but inflation is fuzzy in a way that pressure in physics is not. Economists would not draw such graphs if they did not yield useful information.

Now, in informal parlance we talk about wages and prices differently. We talk about wage-price spirals, for instance. Even some economists talk about wage-price spirals, I believe. When non-economists talk about inflation, they think of price inflation, not wage inflation. IMO that is not a fallacy, that is another way of dealing with the non-randomness mentioned above.

Min, I agree with what you said, but I am really thinking of two criticisms:

1. Yes, economics is more fundamentally fuzzy/subjective than the natural sciences, however I feel that many economists (not necessarily those here) refuse to acknowledge this. Instead many try to ape physicists in the way they go about the subject, and from many we get ideological, simplistic decrees as if they have stumbled onto absolute truth. Moreover, we get official bodies making predictions they have no right to be confident in - when the current British government came into a office, a rash of groups like the IMF, as well as the newly set-up government body the OBR made various specific claims about how growth was going to bounce back. All have since scrambled to reduce their predictions as the data has got worse. To my mind, I can't see how I non-biased person would have given their original figures any credence - they didn't seem scientifically driven. Personally, I'm thinking that the inherent "fuzziness" of economics means predictions need to be more qualitative rather than quantitative. In fact, this is related to my skepticism of NGDP targeting - it seems too technocratic. The concepts were dealing with seem too intangible for me to truth this can be solved adequately by the CB just pulling a lever.

2. Paul Krugman likes to talk about a "Dark Age" of macroeconomics, but I might go further and say there never really has been an enlightened age (in both macro and micro). It's understandable based on the "fuzziness" of the social sciences that you will get a lot more disagreement, but I do believe that some of it is because economic is in a particularly bad way. There seem to be disagreements amongst experts on the most fundamental basics of economics. There should be at least a bedrock of common ground which the vast majority agrees on, but that doesn't seem to be there.

"The concepts were dealing with seem too intangible for me to truth this can be solved adequately by the CB just pulling a lever"

should be:

"The concepts WE'RE dealing with seem too intangible for me to TRUST this can be solved adequately by the CB just pulling a lever"

"There should be at least a bedrock of common ground which the vast majority agrees on, but that doesn't seem to be there."

The vast majority of economists agrees on a number of issues e.g. the elimination of trade barriers, the desirability of effluent taxes, the ill-effects of rent controls etc. The fact that they are generally ignored on these issues suggests that economists really aren't as influential as some people think.

I think most economists are perfectly aware that they are doing social science and I think it's incorrect to claim that most economists are trying to ape physics in any way. The latter is a popular meme in certain philosophy of science circles though, due perhaps to some rather incoherent work by Phllip Mirkowski.

Where apeing does become a problem (particuarly, oddly enough, in Mirkowski's work) is in the assumption that economies are deterministically predictable, hydraulic, and simple mechanical systems. However, the science aped is not physics, but engineering. In truth, economics at its best is more like horticulture than engineering.

Anyone who gave IMF forecasts no credence would probably have a bias.

The purpose of NGDP targeting is precisely to stop what happens with qualitative targets: the institution does what is in its best interest to do and has no hard rules constraining it. Looking at US monetary policy in recent years, qualitative targets ("keep inflation and unemployment low") have not been a success that inspires imitation.

The enlightened age of macroeconomics began with Friedman 1968 and ends sometime in the late 1970s with the rise of the New Classical and New Keynesian schools. For a brief period, a majority of macroeconomists neither thought of people as being ridiculously stupid nor as being ridiculously intelligent. These days, we have to look hard (to places like this ;) ) to find such macroeconomists.

"Due to the way inventory and depreciation are accounted for, at higher rates of inflation, real corporate taxes increase"

This is a very strange idea. First, companies can choose between FIFO and LIFO for inventory accounting. Second, it assumes that price of depreciating assets do not rise. Thirdly it assumes that companies make the best capital allocating decisions of all because households are not allowed to depreciate anything. Well, the problem is that it is households and only households who validate any capital allocating decision.

A thought: is NGDP targeting a polite way of saying "prices and incomes policy"?

I appear to have lost a comment. To try again: Nick's point is that inflation, strictly speaking, is generalised inflation - a general rise in prices, including the price of labour. Therefore, workers shouldn't worry about inflation per se, because it's only inflation if it's generalised inflation. And, to operationalise the point, NGDP targeting specifically aims at generalised inflation in the current situation.

I don't have a problem with this, and I don't really know how Nick has got the impression I'm some sort of Bundesbank-minded deflationist, especially after I defended the Phillips curve in another thread.

My point is that there is no magic wand that adjusts all prices across the economy, frictionlessly, instantly. NGDP targeting would probably work well if it was possible to dial-in an inflation rate of x% on all wages and prices directly. Instead, it proposes to act through the financial economy. In the analogy someone used about horticulture, it proposes to spray the whole garden with fertiliser.

Typically, we assume that inflation is generalised because wages are bid up to match prices. What else could possibly be happening to equalise the two? What do *you* think is happening to make this work? Even if we tell a monetarist/ratex story instead, the rational expectations involved would be those of the people on both sides of the table negotiating wages.

Otherwise, you have to assume that firms respond to rising prices by putting up wages for some exogenous reason. It can't be that otherwise competitors would hire their staff, because we're in a recession and there are millions unemployed. I'm really struggling to imagine micro-foundations that *don't* involve wage bargaining - an economy of co-ops where increased nominal earnings feed straight into the payroll? an economy of altruists?

It's not the end-state of generalised inflation I'm concerned about - it's the process of getting to it.

Alex,

"A thought: is NGDP targeting a polite way of saying "prices and incomes policy"?"

I'd say it's a recognition that the best way to stabilise aggregate demand is to stabilise aggregate demand, rather than stabilise aggregate demand OR destabilise it in order to hit an employment/inflation/gold price/exchange rate target.

In the UK at the moment we have a situation where wage inflation is less than CPI/RPI. Real wages are falling.

If I am an employee, all I care about is my real wage. If my wage is going up at a rate slower than the rate at which the price of stuff is going up that makes me sad, because I can now buy less stuff than I used to. That is what happens to people when real wages fall. It is what is happening in the UK right now.

I'm mystified by Nick's discussion of real wage income. Why do I care about the fact that falling real wages might, given the right elasticities, cause employment to rise, leading to an increase in real wage income? I do not, in fact, care. I am poorer than I was. So is everyone else. This is because prices are going up faster than wages. That is why we are poorer.

Here's a big part of the story which Nick left out: time. Some of the time I run a drinks business. When I take on an employee, that employee negotiates a salary with me, and the custom in the western world is that salaries are quoted and fixed in units of currency per year. When I'm not negotiating salaries, I'm negotiating with wholesalers, distributors, and other customers. The prices here are determined by whoever did better during the negotiation arm wrestle. It's a tad more complicated than that: prices to our wholesalers are fixed over a year, prices with individual accounts may be negotiated through the year. But you get the picture: wages fixed, prices volatile.

Let's think about what happens when one of my employees walks into her local greengrocers. She sees that the price of apples has gone up a bit. Here is how she should correctly interpret what she sees. She should correctly say: "I see that the price of apples has gone up a bit." She is correct to say this because her wages are fixed, and prices aren't, and she can now afford to buy fewer apples. Inflation has made her poorer. And in that respect, she's just like everyone else in the UK right now.

Alex: "A thought: is NGDP targeting a polite way of saying "prices and incomes policy"?"

No, it's very different. In NGDP targeting the central bank moves the AD curve to get the target level of nominal income where AD intersects AS. In prices and incomes policy, the government passes a law against price and wage increases, to stick a third curve into the diagram, so try to prevent the economy going to where the AD and AS curves intersect.

"I appear to have lost a comment."

We have been having trouble with the spam filter. Sorry about that. But I looked in the spam filter and couldn't find it (I did find one comment there by Jacques Rene). My guess is that it got delayed, for some strange reason, and it's the comment that appears slightly later. Again, sorry about that.

Let me try this: suppose, just suppose, that wages were set by supply and demand for labour. The demand for labour is some function of *real* wages Ld= D(W/P) because if all prices double, and all wages double, that would have no effect on the quantity of labour demanded. And the supply of labour is also some function of real wages Ls=S(W/P) because if all prices double and all wages double that would have no effect on labour supplied. Setting demand equal to supply we get L=D(W/P)=S(W/P) as the equilibrium. So, if we start in one equilibrium, a doubling of all prices will cause a doubling of wages and no change in employment.

Now, you might say that wages are not set by supply and demand in a competitive market. And I would basically agree with you. But the same point still stands. Wages will be set in some general way depending on firms' behaviour and workers' behaviour. But if firms' and workers' behaviour depends only on *real* wages (and it should, unless one or the other suffers from some form of money illusion), you still get the same result.

None of this means that *other* shocks can't cause real wages to change. And none of this means that monetary policy doesn't have real effects in the short run, because prices and wages and expectations don't all adjust instantly when monetary policy changes the target rate of inflation.

"My point is that there is no magic wand that adjusts all prices across the economy, frictionlessly, instantly."

Totally agreed. That's why monetary policy really matters. Different monetary policies will have very different short run effects. That's why getting monetary policy right really matters. If it were only the long run inflation rate we were talking about, and we went to the long run instantly, monetary policy wouldn't matter that much.

"NGDP targeting would probably work well if it was possible to dial-in an inflation rate of x% on all wages and prices directly."

We think it's exactly the opposite. We think NGDP targeting works well precisely because it is not possible to dial in an inflation rate of x% on all wages and prices directly. If we thought that all prices and wages did adjust instantly, all of us NGDP targeters, and most of the economists who disagree with us, would decide that this whole argument was a complete waste of our time. You could target NGDP, inflation, the price of gold, the price of foreign exchange, whatever, and it would make damn all difference to booms and recessions. Me, Scott Sumner, and most of the other monetary economists, would find something else to do with our time. I would become a microeconomist, or spend my time fixing cars.

This debate with Alex has been really instructive.

I don't really know anything much at all about Alex, other than that he blogs at Fistful of Euros, which is a good blog. I am quite confident that Alex is much more intelligent, much better educated, and much better informed than the average person.

But it is quite clear from the above comments that there is a total disconnect between Alex's views on what macro/monetary policy is all about and economists' views.

Let me lay them out:

Most of us believe that prices and/or wages and/or expectations do not adjust instantly to changes in aggregate demand. And because of that, different monetary policies will have very different effects in the short run. Some monetary policies will be very bad, because they will cause (or allow) big fluctuations in aggregate demand which will cause big and bad fluctuations in employment and real output/income. And all the arguments between keynesians and monetarists and their like over the last 70 years, between inflation targeters and NGDP targeters, and those who just want the central bank to do what it thinks best at the time, and everyone else, have been over how best to do that.

And most of us believe that in the long run prices and wages and expectations eventually do adjust, and that, if we ignore all those very important short run fluctuations, about the only thing that monetary policy affects in the long run is inflation. And we argue a bit about the long run costs of inflation, and most of us think that having inflation below 2% in the long run might be unwise, because of various zero lower bounds And that there aren't any advantages of going much above 2% in the long run, and there are some costs too. So whether you decide on an inflation target or an NGDP target or whatever on short run grounds, it's probably best to have a monetary policy that delivers somewhere around 2% inflation on average in the long run.

A minority of economists disagree, think that prices and wages and expectations do all adjust very quickly, and think that the short run effects of monetary policy are just like the long run. They either think that monetary policy really doesn't matter very much, or else have a hard time explaining why it matters.

if all prices double, and all wages double, that would have no effect on the quantity of labour demanded. And the supply of labour is also some function of real wages Ls=S(W/P) because if all prices double and all wages double that would have no effect on labour supplied

You're still not addressing the point. You haven't actually addressed the point in the whole argument. You just repeat that everything is fine so long as "all prices double and all wages double". Well, yes. I do not disagree with this and have never done. In fact, I find it obvious and even trivial, and I can do without a repeat of A-level economics class.

Your logic is perfectly circular. Everything is OK because inflation is defined as a rise in all prices and wages. Why is it so? Why, because inflation is defined as a rise in all prices and wages? What is the mechanism that links them? Inflation is defined as a rise in prices and wages! Why is it so? And so on and so forth.

Actually, this discussion has been completely uninformative. As far as I can see, we're arguing completely at cross-purposes. You seem to have filed me as "rather dim hard-currency type" very early on and to have implemented your working heuristic for this case, before moving on to the script for "difficult student, pull rank and make them cry", and not to have noticed that I'm not at all opposed to inflation...in fact, if you read AFOE with any attention you'd notice that I've been arguing for an inflationary solution to the intra-eurozone imbalances since April, 2010!

PS: But it is quite clear from the above comments that there is a total disconnect between Alex's views on what macro/monetary policy is all about and economists' views.

You're damn right!

There's been some progress. Nick began by pretending that inflation didn't matter. He is now at least admitting that inflation can matter in the short run. And there's an inkling he recognises that part of the reason is asynchronous contracts. Perhaps if he can find a way to introduce the most asynchronous of all contracts, namely debt, into his models, he'll understand that the duration of the short run can be really quite long...

Oliver: God give me strength. "Admitting"!!!!I've been playing around with things like asynchronous price/wage setting and the short run non-neutrality of money for the last 35 years. Why do you think I'm interested in things like NGDP targeting, for heaven's sake!

One of the problems with this whole discussion has been that your opening post should have begun: "If you assume a closed economy with instantaneous adjustment, oh and by the way we're going to absolve ourselves of the need to think about stock flow consistency, then the following belief about inflation is fallacious..."

If you admit asynchronous contracts, then the fallacy you described at the outset starts to look a lot less fallacious. Depending on the type of contracts, their duration, how easy to unwind etc., then in certain circumstances there need never be adjustment between changes in the price and wage level. Imagine a highly indebted household sector, with a constraint in the form of a maximum acceptable debt:income ratio. Suppose also that a lot of these households are struggling (which is why the sector is indebted). "Struggling" means that some of the time consumption = wages, some of the time consumption > wages, which is how debt accumulates.

Prices rise (this is an open economy, with lots of essential imported goods, and price movements are independent of the state of the domestic economy). Now what? If households are already at their debt:income maximum they can't maintain their standard of living by borrowing more, so they reduce their real consumption. Since everyone's consuming less, real wages start to fall. The household sector now has to service its debt burden, which is fixed, out of declining wages, so it consumes less...

It always surprises me when people assume that debt + inflation is good news. They only think this because the assume, wrongly, that price changes and wage changes move in step. They don't, and when the debt overhang is significant, the effect of that mis-timing is pernicious.

I have no idea, by the way, what the duration of short run in the scenario I've just described is.

Try explaining why inflation is bad to all the homeowners that had mortgages whose principal and payments dropped to insignificance over the life of the mortgage as their wages and the value of their house inflated.

Or all those people who kept refinancing and pulling the inflation equity out of their houses to use in place of income.

Oops, it worked for 70 years until the housing inflation turned negative.

Borrowers like inflation. Lenders like deflation until the borrowers can't make the payments anymore.

Inflation, whether positive or negative, requires adjustment. If it is slow, most people don;t mind too much. If it is fast, then it may not be possible to make the adjustments fast enough. Or you may spend all your time making adjustments rather than doing useful work.

Scientists, and probably economists, like to make the quasi-static assumption because it makes the math much easier. Some times, in real life, things move just to fast for the quasi-static assumption to hold.

Then of course there is the all other things being equal clause. Rarely are all other things equal.

If this is not emphasized to the student, then the student is left with the question as to why all these models are so valued by economists even though real life does not act that way at all.

And here, with perfect timing, come those Econ 000 know-nothings from the UK's Office of Budget Responsibility with their Economic and Fiscal Outlook for the UK:

"The contribution of nominal consumption to money GDP was only 0.1 percentage points lower than forecast, suggesting that higher-than-expected inflation from the second half of 2010 was a key driver of the weakness in real consumption. This was largely the result of unexpected rises in energy prices and global agricultural commodity prices."

p. 28, here: http://cdn.budgetresponsibility.independent.gov.uk/Autumn2011EFO_web_version138469072346.pdf

Mystifying.

Oliver: "Nick began by pretending that inflation didn't matter."

No, he didn't. That's what you thought he said and I can see why you thought that, but that's not what he meant at all.

Nick was actually only identifying a particular argument as fallacious. He was saying that 10% inflation does not make all goods 10% more expensive.

Sure, it could make everything 10% more expensive for a little while, but rather quickly sellers would notice that no one is buying anything because everything is suddenly so expensive. At that point, most prices will rapidly adjust. Some will not adjust rapidly, for various reasons. Wages will probably adjust relatively slowly. Because of that, in the very short term, purchasing power (for many/most people) will fall, but that decrease will not be 10% for very long (maybe a day?). Because various prices (including wages) will adjust at very different rates, relative prices will rapidly change. That can cause all sorts of negative effects (and maybe a few positive ones). Does anyone disagree with my description?

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad