Sunday, November 28, 2010

Start eating out more?

Economic models take it as a given that, in the face of expected inflation, savers start to spend more.

In practice, this isn't at all clear. I keep posting a simple example, and nobody seems to have a good story for how the actions taken by the individual in the scenario will lead to broad CPI increase.

Here's the example again -- please post your mechanisms in the comments:

You're a responsible Brazilian living in your decent Sao Paolo apartment (paid off!). You have a tidy pile of cruzeiros in your local bank, saved from the income your reasonable private sector job generates. But it's 1979 and you're worried about inflation looming on the horizon. What do you do?

I'm interested in two things, but all ideas are welcome. First, the above situation is not hypothetical, so people actually made decisions in these circumstances. If anyone knows what folks did in real life, I'd love to hear them. Second, this example is meant to focus on exactly how increasing inflation expectations actually leads to rising CPI. No hand-waving please, I need transactions!

Labels:

57 Comments:

Blogger J.P. said...

What about shifts in what assets our hypothetical Brazilian chooses to hold? Buying either durable physical commodities or forex could offer a better real rate of return than holding domestic currency. If the cost of durable physical commodities increases, then some of that would be passed along to final consumer prices. Likewise, a depreciating exchange rate would raise the portion of CPI that measures the cost of foreign imports.

3:37 PM  
Blogger Scott said...

I lived in El Salvador in 1979, and I can tell you what ALL wealthy Salvadorans were doing with their colónes: they were wiring them off to Miami as quickly as they could where those funds could sit safely (unspent) in dollar-denominated accounts. I suspect that's what any sane Brazilian with notable savings was doing in '79 as well.

10:49 PM  
Blogger Игры рынка said...

I lived through a hyper-inflation following the Soviet Union collapse. And it had supply-shock reasons. The first thing people do in the face of coming inflation is get dollarized. Volumes of purchases of any "retail" stuff definitely do not increase as people simply and constantly switch in and out of dollars whenever they need to purchase anything above a basic daily expense or people
just transact in dollars.

People stock stuff during supply-shock not because they expect prices to increase but because they are not sure whether they can buy what they need (e.g. salt, sugar) when they need it. Future local prices are irrelevant as prices of all goods are benchmarked against dollars. However this "consumption" behaviour can make demand for goods very volatile. And with sticky prices it has its effects on inflation.

No reason to expect economy to dollarize due to demand-pull inflation.

2:01 AM  
OpenID vimothy said...

Winterspeak,

You're just making restrictions that are bound to give you the answer you want to get. If it doesn't increase the MPC it is not going to add to inflationary pressure.

I too doubt that a rich Brazilian is going to want to speand all of his or her accumulated wealth on consumption at the first sniff of inflation. FWIW.

7:40 AM  
Blogger winterspeak said...

JP: which durables? Can you be more specific?

vimothy: what restriction am I making? I'm picking a perfectly reasonable example that happened in real life! And my guy isn't even a rich Brazilian, he's a middle class Brazilian.

Here's a guy with some limited means, who has just increased his inflation expectations. Please tell me what you think he will do, and how that will feed into measured increases in CPI.

This is the core conduit claimed by the "inflations expectation" crowd and I have yet to see a single actual mechanism. Do you have one?

8:58 AM  
Blogger Rogue Economist said...

If you are worried about looming inflation, you're less likely to eat out. You need to preserve your diminished purchasing power for the basic necessities. You will trade down to less costly substitutes. You will repair stuff rather than buy new ones.

4:53 PM  
Blogger winterspeak said...

Rogue: Shhhh. Don't let an economist hear you say that. It's opposite to what their models demand.

Also, don't think about what you might do if your expected real return on investment falls as well.

9:08 PM  
Blogger Scott said...

"If you are worried…"

You could have stopped right there, Rogue Economist. You get the psychology just right. Any economic uncertainty – even looming inflation - leads to worry. And worry will almost always drive businesses and consumers to curtail spending and preserve capital.

"I just can't spend this money fast enough" is something you will never hear from worried economic actors.

10:20 PM  
Blogger winterspeak said...

Scott:

Periods of inflation and hyperinflation do happen, and households do position themselves for it. There have been periods where economic actors will say "I just can't spend the money fast enough".

The issue is that they do not position themselves the way standard macro models demand.

This entire post is all about what households actually /do/ in response to an inflation expectations shock.

9:24 AM  
Blogger Ulisse Di Bartolomei said...

The Fiat patent fraud. About the Fiat hybrids: the technology double clutch with electric motor between has been stolen by a patent that Fiat Company has never wanted to purchase, but only shamelessly to copy. This hybrid solution will be the basic technology with Chrysler's electric and hybrid car program. Please give a look in my blog where the "vitality" and boldness of the Fiat planners it appears in all of evidence: http://dualsymbioticelectromechanicalengine.blogspot.com/
If the industries can afford unpunished to copy the ideas and defending it need very expensive trial, to which target need the patents? How to defend the rights of private inventors? How our young people can find intellectual courage if the economic potentates crush the rights of the single ones? Whoever is about to ask for a patent or wants to propose a proper patent to a big firm I suggest to give a look to my experience with the Fiat, to get able to operate with better adroitness. Thanks and good time to everybody. Ulisse Di Bartolomei

11:55 AM  
Blogger Scott said...

Winterspeak: Okay, I should never have said "never." It was late, I was grumpy.

Still, I think the "spend it as fast as you get it" scenarios are bogeymen to an extent - like scary stories inlationistas tell their children around the campfire. Uncle Otto used Marks as kindling! This could happen here!!!

4:24 PM  
OpenID gbgasser said...

Good post Winterspeak. Excellent job at getting at the nub of the argument.

Truth be told, I'm always worrying about inflation, even now, not in the manner that the macro models say but I'm always worried that what I need now may not be available at todays price when I decide to purchase it later. Now if its truly a NEED I dont wait, if its something I want but there are other priorities I always am wondering if next month I'm going to regret waiting. Prices change, and life becomes more expensive the longer we live. However I dont consider it to be the priority of our macroeconomy to keep enough people unemployed so that my inflation adjusted dollar today will by what it could have bought 5 yrs ago, as it seems many people do.

One thing that we need a national conversation about is the fact that the rest of us dont need to adjust our lives so the almighty savers will be able to get the equivalent of that which they "forwent" consuming 10-20 years ago. Because lets face it, thats what this whole inflation discussion is about, the savers (who supposedly "produced" more than they needed and lent it to the rest of us) want to be sure that 20 years from now they can get (at least)the equivalent of what they chose not to consume in the past. I say (mostly) screw you! You want something get it now. If its not available invest in that which will make it available or STFU. You dont get to make demands on the rest of the workers so your future consumption desires arent inhibited.

12:05 PM  
Blogger winterspeak said...

gbgasser: I wouldn't blame the savers. They are fundamentally not the ones causing unemployment, any more than it's wet pavements that cause rain.

The government can easily make all of this unemployment go away if they just did their job. And academic economists, with their financially illiterate models, are the ones who are giving the government bad advice.

4:33 PM  
Blogger Greg said...

I understand what you are saying and I agree. However dont you think the reality of the situation is that those who have accumulated much have "bought" control of economic policy through think tank stuffing and politician buying and that this is driving policies which have to protect those valuable savers?

Look, I like to save as much as the next guy but my saving is not the engine of future growth, its the result of past new investment, my saving didnt come from someone elses saving, it came from someone elses spending!

7:29 PM  
Blogger winterspeak said...

Greg: I think the reality of the situation is that academic economists don't understand monetary operations and think double entry bookkeeping is beneath them. Therefore, the policy makers they are or influence have no idea how to respond to a savings desire shock.

And no, I don't think Paul Krugman has been bought by those who have "accumulated much". He just doesn't know how any of this stuff works.

Savings is the account of record for investment. Demand is what drives investment. The rain is cold and wet, we have an umbrella, but we don't open it because Krugman can't tell and asset from a liability. I'm not blaming the weather.

10:05 AM  
Blogger Greg said...

Let me try one more way of asking this.

Is it not true that in our system the amount of influence you have is directly related to the amount of money you have?
Is it also not true that you will use your influence to benefit yourself, often times even if it is not beneficial to your community as a whole?

If the previous statements are true then accumulations of savings will lead to accumulations of power and influence unless an outside force capable of attenuating these forces acts in a way that mitigates the power.

Now, how the outside agency accomplishes this matters but if the outside agency continues to agree that the level and value of these savings is what is most worth defending, the initial distribution of the savings power cannot change.

I'm arguing that having too much of the savings in too few hands is problematic, not that saving is the problem. How would you argue that I am wrong?

Placing limits on savings, taxing accumulated wealth above a certain amount are options (not sure I agree with them because there might be better ways).

More to the content of your post though, its clear to me that these hypersavers are more and more likely to continue accumulation if they think inflation is imminent. They will act to mitigate inflation and therefore NOT spend.

8:25 AM  
Blogger winterspeak said...

Greg:

I think the ability for money to buy influence peaked during the Robber Barron era that ended in the 1930s, although their legacy lives on through their Foundations. But those once great estates have just become repeaters for the Academy -- can you see any light between the Ford Foundation and the Harvard Social Studies Department?

So I agree that being rich buys you political influence, but there are other ways to have influence, and those are much more relevant to today's world. Paul Krugman is not in the pocket of the Koch Brothers, and Paul's why we have 10% unemployment right now, not Charles or David!

I don't see how "placing limits on savings" or anything else you suggest will make Paul Krugman understand the economy, finance, and monetary operations. And this is the crux of the problem.

Moving on to content, I think it's clear in this thread that there is no mechanisms by which a positive inflation expectation shock can turn savers into consumers. It may change the form of the savings though. There's on exception to that, though, which I think I'll describe in my next post.

2:00 PM  
Blogger Nick Rowe said...

Hi Winterspeak:

You want "transactions"?

And you want those transactions to be consistent with double-entry bookkeeping?

The simplest story has no change in transactions whatsoever. So absolutely nothing happens in the double-entry books, except that exactly the same amounts of goods and labour and everything else get bought and sold at higher prices.

And that simplest story is the essentials of the answer. Any change in actual transactions, though realistic, given short-run non-neutrality of money, is just a complication. It's not the story. It's not the reason the CPI rises. It's a side effect of the story.

So, if I told you a story in which there were zero change in actual transactions, merely a change in the prices at which those same transactions were made, would you reject it out of hand, for that very reason?

Let's get to the guts of the issue right now.

The guts of the issue is the distinction between quantity demanded and quantity actually bought. And between quantity supplied and quantity actually sold. Because, as the accountants remind us, quantity bought is the same thing as quantity sold. But, as the economists remind us, that does not mean that quantity demanded is the same thing as quantity supplied. It is prices, and therefore indices of prices like the CPI, that adjust to equalise quantity demanded and quantity supplied.

Everyone wants to get rid of money, because it wont be worth as much in future. At existing prices, that creates an excess demand for some other things (no, it does not matter what). A rise in the price of those things, caused by that excess demand, would cause people to substitute into still other things, creating an excess demand for those things. Etc. So, next morning, all the accountants look at their books and see no absolutely no change in the volume of transactions on their books. Just a proportional increase in the prices at which all those transactions took place. Still the same 100 pairs of shoes bought and sold, but at $20 each, not $10 like yesterday.

3:30 PM  
Blogger winterspeak said...

welcome, nick, and thank's for representing the "immaculate conception" establishment position. You're begging the question though, so I'm going to press you on one point. You say "Everyone wants to get rid of money, because it wont be worth as much in future. At existing prices, that creates an excess demand for some other things (no, it does not matter what)"

Please list 10 items that my guy in Sao Paolo -- who has been waiting for weeks now with his pile of cruzieros!! -- will buy once he expects there to be inflation in his future. Enough handwaving -- give me a list. And make it a good one or I'll declare victory ; )

4:11 PM  
Blogger Игры рынка said...

yes, and please clarify why quantity sold should have a structural break from one day to another to allow for a CPI adjustment. And also how the gap between quantity supplied and quantity sold gets resolved.

1:29 AM  
Blogger Nick Rowe said...

Thanks Winterspeak!

"Please list 10 items that my guy in Sao Paolo -- who has been waiting for weeks now with his pile of cruzieros!! -- will buy once he expects there to be inflation in his future. Enough handwaving -- give me a list. And make it a good one or I'll declare victory ; )"

That's a really tough one. I don't know anyone in Sao Paolo. I've never even been to Brazil!

Could I answer instead for someone I know (me) living in Chelsea, Canada?

And remember, this is not a list of what he "will buy". It is a list of his additional *quantities demanded*, which is not the same as what he "will buy".. He won't *actually buy* any of these. Because the increase in quantities demanded, by everyone, will cause prices to rise until he stops demanding them.

1. A new, lighter weight canoe. Swift Mattawa, in carbon fusion. Better buy it now, before prices rise, rather than waiting till I'm too old to lift my heavier one. (Do you really want the exact model and manufacturer?)

2. Better cookware.

3. Many crates of Scotch.

4. A new field drain under the back yard.

5. New laptop for one daughter.

6. New stereo for the other daughter.

7. Small farm 30 minutes north.

8. shares in oil/gas drilling company.

9. equity mutual funds (have to ask my broker which).

10. One year's leisure (unpaid leave from work, since the money won't be worth as much, and it would be better to take the leisure now and retire a year later).

Short Market: "yes, and please clarify why quantity sold should have a structural break from one day to another to allow for a CPI adjustment."

??? I'm saying that quantity sold will *not* have a structural break. It won't change at all in my simple story.

"And also how the gap between quantity supplied and quantity sold gets resolved."

?? By the current price level rising, of course. It rises relative to the expected future price level until the expected inflation rate, and real interest rate, are restored to their previous equilibrium values.

10:40 AM  
Blogger Игры рынка said...

Nick, if economy is steadily producing and consuming 100 laptops per year then 100 is quantity demanded, bought, supplied and sold. Why and how should the price of laptop increase?

You say that suddenly quantity demanded will increase (not stochastically but structurally) but quantity supplied will not therefore price will rise to balance the system at the quantity sold which remains at 100. But you give no reason as to why it should be the case. That is exactly the question and you fail to give an answer. So why should quantity demanded increase and why should quantity supplied fail to catch up.

11:04 AM  
Blogger Nick Rowe said...

Short Market:

I am really unsure whether you understand the distinction between quantity supplied and quantity sold.

But, assuming you do understand the distinction, why ever should quantity supplied increase, when the expected future price level rises? If anything, I would expect the quantity supplied to fall. The seller of laptops would want to sell fewer laptops today, since he expects to get a better price for them in future.

That reduction in supply of laptops, added to the increase in demand, would create an even bigger excess demand for laptops at the current price, and put even greater upward pressure on the price.

2:33 PM  
Blogger Scott said...

Nick Rowe: "He won't *actually buy* any of these. Because the increase in quantities demanded, by everyone, will cause prices to rise until he stops demanding them."

Inflation without actual buyers. So, prices rise because wishing (in the aggregate) makes it so? How exactly do you measure an increase in "quantity demanded" without any transactions?

Nick Rowe, again: "If anything, I would expect the quantity supplied to fall. The seller of laptops would want to sell fewer laptops today, since he expects to get a better price for them in future."

I'm not an economist, but I am a businessman. If I were selling laptops I would want to sell as many laptops as I possibly could TODAY - at today's price. Next week's price only matters next week. And I would love any competitor who decided he wanted to sell fewer laptops this week. That's the kind of competitor I want - a total idiot.

Businessmen <3 transactions - today, tomorrow and always!

How does your theory account for "irrational" laptop sellers like me?

10:13 PM  
Blogger Игры рынка said...

Nick, you are lost in microeconomic theories which fall prey of fallacy of composition.

You still fail to explain why (future) price should increase. Or why current demand should increase. And why current supply can not increase.

11:00 PM  
Blogger Greg said...

What Nick is describing is a market that many suppliers WISH they had. Which is not free but monopolized. He is assuming that all laptop sellers will collude I suppose. Maybe thats a valid assumption but it is exactly why we need some outside agency to prevent it. Unfortunately our outside agencies get owned and end up allowing it. What to do?

8:04 AM  
Blogger Greg said...

BTW Winterspeak, great challenge.

I hope we get hundreds and hundreds of responses here from all over the internet. Lets put out some invitations to the Austrians. They never have an inflation story they cant describe, complete with CB boogeymen and wheelbarrows full of multi zeroed paper notes.

8:09 AM  
Blogger Nick Rowe said...

Scott: "How exactly do you measure an increase in "quantity demanded" without any transactions?"

Have you ever visited an auction?

"I'm not an economist,..."

Clearly. But you could try reading a first year economics textbook, on supply and demand.

Short Market: "Nick, you are lost in microeconomic theories which fall prey of fallacy of composition."

Nonsense. I was doing macro analysis. And yes, I do understand fallacies of composition, and the difference between changes in relative prices and nominal prices.

"You still fail to explain why (future) price should increase"

Winterspeak *assumed* there was an increase in the expected future price level. Of course I didn't explain it. I was granting Winterspeak his/her assumption.

"Or why current demand should increase."
Winterspeak never asked me to. That wasn't part of the deal. Do you want me to? Do I need to explain why, if you thought Scotch would be twice the price next year (other things, like nominal interest rates, constant) you might want to buy an extra crate of Scotch this year?

"And why current supply can not increase."

Yes I did. Didn't you read my previous comment? Didn't you understand it? It will fall, if anything, for the exact same reason that demand increases. Tell me, if you were planning to sell some scotch now, and then thought that the price would double next year, would that make you more or less likely to sell the scotch this year?

(Would you short a commodity you thought were going to rise in price?)

Greg: "What Nick is describing is a market that many suppliers WISH they had. Which is not free but monopolized. He is assuming that all laptop sellers will collude I suppose."

Oh dear. No, I was assuming the exact opposite. I was assuming a perfectly competitive market. That should have been obvious, from anyone reading what I wrote, and my talking about quantity supplied and quantity demanded.

You just blew the one chance you had to make a serious critique of my answer. If you had instead accused me of assuming perfect competition, and said that the real world is not perfectly competitive, you would have had a point. And I would have needed to switch to a different, more complex, explanation of why the increase in demand causes prices to rise.

Winterspeak: look, sorry, but your commenters here just aren't up to your standards.

4:32 PM  
Blogger winterspeak said...

Nick: Hey -- comments are still an experiment here. I'm very glad you came, and welcome your contribution. You don't know it yet, but I think they are worth their own post.

Anyway, I try hard not to censor anything in comments. You takes what you gets.

I wanted to pick a realistic example, and thus chose cruzeiros in the 70s. There was real hyperinflation shortly thereafter, and the cruz is no more. I was looking to see what a foresighted individual might do, and how his actions might feed into hyperinflation. I was looking for the self-fulfilling prophesy that seems to lie at the heart of the "inflations expectation" camp.

So although I picked a historical moment when it did happen, I was looking for a reasonable story that help illuminate the mechanism. I did not assume a mechanism, coming up with suggestions for what the mechanism might be was the whole point of the exercise.

9:03 PM  
Blogger Игры рынка said...

Nick, I read your comments and indirectly before but directly now you claim that if producer of oil expects its price to double next year then they will not produce/sell this year. This is what I have trying to clarify and you failed to clarify. Instead you keep on changing the subject, and now blame others, anything to avoid a direct answer.

So is it correct? Yes or no? As simple as it gets.

10:49 PM  
Blogger Игры рынка said...

And by the way, in case you do not know it, shorting a commodity has nothing to do with demand or supply of it.

This you can not learn in your macrotheories. This is real life.

10:52 PM  
Blogger Scott said...

Wow, Nick, didn't mean to trip your "suffering-fools" wires.

My questions were serious. Remember that from school - there are no dumb questions? Maybe that changes once you hit grad school. Your answers (to me, at least) were dismissive and unhelpful.

So, I ask again: how would you, professional economist, recognize/measure an increase in "quantity demanded" without data from actual sales/transactions?

(Most goods are not sold in auction-like markets, no? Where can I see the bids for laptops or milk or eggs or automobiles?)

How would I, laptop selling businessman, recognize an increase in quantity of laptops demanded (without sales increasing) that I could take profit-maximizing advantage of?

And what if my businessman's idea of "profit-maximizing" is the exact opposite of what you suggest, and I work to increase supply in the immediate term to meet this growing demand - and do it fast before anyone else beats me to it?

These aren't trick questions. These aren't veiled critiques. I'm genuinely curious how your model explains the decisions people like me would make in these circumstances.

1:43 AM  
Blogger Greg said...

Nick

You describe a market where a seller of something forgoes sales at X today because he wants to wait and sell at 2X some time in the future.

How does this happen in a perfectly competitive market? You seem to think there is a way that a business, all businesses would just know this. If the businesses would know this then the consumers would likely "know" this too and then prices would rise immediately.

I KNOW this can happen in a monopolized market, by what mechanism does it happen in a non monopolized market?

2:16 AM  
Blogger Nick Rowe said...

Winterspeak:

No worries. And I should say, in praise of your commenters, that they are definitely civil. (I was perhaps a little less so, sorry, but in my defence, there is something about the set up here that raises my suspicions that this is all a trap to lure in some poor unsuspecting MMTer (Mainstream Monetary Theorist ;-)) like myself. And so all my defence shields are on full-alert!)

As for the mechanism, the one I was (implicitly) assuming in my story is that the increase in expected inflation reduces the real interest rate for any given nominal interest rate (whether the nominal interest rate would in fact stay constant is another story, depending in part on what the central bank does).

A reduction in the real interest rate:

1. Reduces the relative price of current to future consumption, and so causes a substitution effect towards demand for current consumption and away from future consumption. This could apply to restaurant meals (just as you will demand more restaurant meals on Wednesdays if they have lower prices Wednesdays). But it would have a bigger effect on consumer durables, which are really closer to investment goods than consumer goods. That's why most of my examples are consumer durables.

2. The reduction in real interest rates, via exactly the same relative price substitution effect, increases the demand for investment goods, whose rate of return is now relatively more attractive. That's why I included the farm and the oil shares and equity mutual funds.

So, with desired consumption and investment demand increased, and assuming demand=supply before, there's now excess demand, which pushes prices up.

That's basically the textbook answer to your question. (Except I have glossed over the central bank's reaction function, but maybe, in Brazil at that time, my implicit assumption is not far off the truth).

4:18 AM  
Blogger Nick Rowe said...

Continued (my comment was too long):

Short Market:

Start in equilibrium, Suppose everybody in the economy expects all prices to double next year. The oil producers will reduce quantity supplied at current prices (his supply curve shifts left). But at the same time oil consumers will increase quantity demanded at current prices (their demand curve shifts right). The same happens across the whole economy (we are doing macro, and must avoid fallacies of composition). Current prices immediately double in response to this excess demand for oil (and everything else). The increase in current price causes a movement along the new demand and supply curves (increasing quantity supplied and reducing quantity demanded) until the economy is back in equilibrium at a higher current price. But with exactly the same level of quantity demanded, supplied, bought-and-sold, and produced, as before.

I have answered your question very clearly and as simply as it can possibly be answered. (My answer does assume that all prices adjust instantly, which is an obviously false assumption, but I can't relax that assumption and keep it simple at the same time).

Scott: "So, I ask again: how would you, professional economist, recognize/measure an increase in "quantity demanded" without data from actual sales/transactions?"

If prices adjust instantly, in competitive markets, so quantity supplied = quantity transacted (bought-and-sold) = quantity demanded, then we cannot observe the three quantities independently, except in special cases like auctions (for example, the quantities bid and offered on stock markets). We can observe them independently in cases where prices are fixed (e.g. rent controls).

Otherwise, they are unobserved theoretical constructs. No apologies for that.

All models abstract from reality. And no apologies for that either.

How does an increase in demand cause prices to rise in real world markets? All real world markets are different. In many (most) markets, the increased demand *would* cause an increase in quantities transacted *before* prices rise. David Hume noticed that 250 years ago. But to explain that better I would need to drop the assumption of competitive auction markets and switch to monopolistic competition with price-setting firms (my preferred model). I would need to draw lots of diagrams, and the explanation would be more complicated. I couldn't do it here. Give me a blackboard, a couple of hours, and people listening who had already sat through ECON1000, and I could do it.

And when I read your comment below the bit I just quoted, I surmise that the (implicit) model of the firm at the back of your mind is also much closer to monopolistic competition. Which is not surprising. In my opinion, monop comp is a lot closer to what most real businesses do than perfect comp. And the very reason I built my first macro model with monop comp in 1987 is precisely because of that. But there's no way I can explain it with just words. I need pictures. And equations too. Sorry.

It's hard enough to explain perfect competition using just words, where I can't even draw supply and demand diagrams to show you what I mean. Damned hard.

Greg: "How does this happen in a perfectly competitive market? You seem to think there is a way that a business, all businesses would just know this. If the businesses would know this then the consumers would likely "know" this too and then prices would rise immediately."

I am assuming (because it's simplest) that *everyone* does know this. And that current prices *will* rise immediately.

And see my explanation to Short Market above. I think that's what you are looking for? If not, I don't understand your question.

4:18 AM  
Blogger Nick Rowe said...

This comment has been removed by a blog administrator.

4:19 AM  
Blogger Nick Rowe said...

This comment has been removed by a blog administrator.

4:22 AM  
Blogger Nick Rowe said...

This comment has been removed by a blog administrator.

4:22 AM  
Blogger Nick Rowe said...

Sorry about the triple/quadruple post. My original comment was far too long, then I screwed up in the cut-and-paste exercise.

Winterspeak: please delete the 3 above. The first two contain everything.

4:34 AM  
Blogger Nick Rowe said...

Nick's night mare exam question to torture economics grad students:

"Using a simple macroeconomic model with monopolistically competitive firms, explain how an increase in the expected future price level will cause an increase in the current price level. You must use words only, with no diagrams or equations. Your answer must assume no previous knowledge of economic theory or concepts on the part of your reader. Try to make your answer as realistic as possible, using 10 real-world goods as examples, that would be relevant to a person living in Sao Paolo Brazil. You have one hour to answer this question. Please write clearly."

4:44 AM  
Blogger Nick Rowe said...

"Your answer will be graded by: an accountant; a businessman; a Russian day-trader; and an economics professor. You must satisfy all these examiners, plus anyone else who happens to wander by, in order to pass the course. And don't forget to be very precise about all the mechanisms involved in what is a highly interdependent system with simultaneous causation".

4:58 AM  
Blogger Игры рынка said...

Nick: "Start in equilibrium, Suppose everybody in the economy expects all prices to double next year."

You must be joking, right? Economics must be really complex (as per Fed analysts) so that even such a simple construct as equilibrium and demand/supply can not be explained in a couple of sentences without violating basic rules of logic. You sentence above logically reads as "true is false".

But lets do a simple reality check. Inflation has been here with us for decades. So if inflation expectations really exist then we should all have them. They might change from now to then but overall they should be positive. How do you explain firms cutting prices and running "sales" if we all know that inflation will make prices higher tomorrow forcing consumer to pay these prices?

Everybody knows that laptop prices will be lower tomorrow but people still buy them today. Everybody knows that oil prices will be higher tomorrow yet oil producers still sell oil today.

Firms care about revenues which depend on two factors: volume and price. Firms do not care about prices. Prices are meaningless for firms. Firms need revenues which they compare to costs. So on the way up the first thing they adjust is volume and then price (and gain on it). On the way down the first thing they adjust is price and then volume (and still gain on it). Firms need revenues because they need to sell. No sales? Then you are bust and out of business. And sometimes they sell at any price as long as it covers variable costs.

Consumers do care about prices because their revenues are given and fixed but they can adjust volumes. So first thing they adjust is volume and then they might accept higher prices. They will not challenge higher prices only for basic necessities.

So go and figure your equilibrium which is not really equilibrium because by simply dreaming ... sorry ... expecting you can move equilibrium. But dreaming is not doing, right? And why not expect lower prices will I ask? Would be a dream world consumers, wouldn't it? No, instead you create an artificial construct called "inflation expectations" and viola. However you simply hate consumers and love firms so prices always move up in whatever model you construct. Yes, all models abstract from reality but GIGO is a very well known concept.

Inflation expectations is a dreamed up macro-concept and sorry you consistently fail to explain why it is not the case. You, supposedly a professional educator, failed your exam.

5:33 AM  
Blogger Greg said...

So Nick

What is the tipping point? How many people with inflation expectations does it take to drive prices up right now?

It seems to me from your story that if we start hearing about people walking into an Athens computer store with $1,000 dollars wishing to purchase a laptop and they are denied, not because they dont have one for 1000$ but because they want to sell it for more next month, then we should start to worry. Has this behavior EVER been observed?

I'd just tell the guy to go to Atlanta and buy one.

Like I told Winter in another comment I ALWAYS have inflation expectations. Inflation is a part of life in a monetary economy. Probably even in a barter economy as well, it would just look different.

6:02 AM  
Blogger Nolan said...

Hi Professor Rowe,

High School Social Studies here with some economics background (the first year econ textbook I read was Samuelson's 2001 edition). Besides the two generic textbook 101 explanations you bring up in the first part of your comment of how lowering the real interest rate impacts the broader economy via the substitution effect, isn't there also an income effect? So yes, reducing the real interest rate may lower "the relative price of current to future consumption," for consumers and "increases the demand for investment goods," for businesses, but it also reduces the income of savers, no? It's true that I only took a handful of economics courses as an undergraduate, but I must say I never really heard a good explanation from any of my professors of why substitution effects should dominate income effects in certain markets or why the opposite should be true in other markets. Perhaps this is addressed more fully in graduate school econ classes.

6:25 AM  
Blogger Scott said...

Nick, thank you for the detailed and thorough responses.

11:31 AM  
Blogger NKlein1553 said...

P.S. Professor Rowe,

The challenge you pose to a hypothetical economics professor above at 4:44 A.M. is faced all the time by us high school economic and social studies teachers who have to create, propose, and explain curriculum maps to school board members and superintendents =)

P.P.S.

I'm the "Nolan," from above. No idea why my first name got published instead of my regular moniker "NKlein1553."

11:35 AM  
Blogger Vivianne Vilar said...

My mum (and all family) used to buy dollars. I can't tell you about 1979, because I was only born 2 years later, but that is what she did from when I was a kid until 1994, when the Plano Real was implemented.

3:00 PM  
Blogger Oliver said...

Would he really be worried about inflation if it hadn't already started? Are you looking specifically for examples where expectations trigger initial inflation, or would you be satisfied if one argues that expectations can exacerbate inflation that is already well underway?

Seems to me the second case is much easier to argue, particularly if seen from the supplier's and not the consumer's point of view. If consumers are already conditioned to expect significant and varying price rises, it seems plausible for suppliers to want to get ahead of the process and raise prices before others do, using uncertainty as a further market imperfection to profit from. I doubt whether uncertainty over prices alone can be strong enough for it to trigger inflation though.

2:17 AM  
Blogger ds said...

I am not sure what you are getting at here. You assume that macro-models use forward-shifting consumption to explain how an increase in inflation expectations leads to actual inflation? I'm not so sure that is what most models would assume. An increase in inflation expectations would just lead to a new round of price setting. In the case of the hypothetical Brazilian, he wouldn't necessarily shift consumption forward or sell his cruzeiros -- he would just go to his boss and demand a raise. If prices are sticky, this raise could temporarily increase the worker's real wealth, which he would then spend. Otherwise, the business would raise prices just as quickly to negate any real effects.


Maybe a good idea would be to take a relatively widely-used macro dsge model and dissect it to get at its fundamental assumptions regarding inflation expectations. It might help to give readers a better idea what you mean when you critique mainstream models.

Very interesting idea and post -- thanks a lot!

7:22 AM  
Blogger winterspeak said...

Oliver: In my scenario, although inflation has not begun yet, the Brazilian decides that it is coming, and takes action with his life savings of cruzieros. This is a realistic scenario as this actually happened, and someone (everyone if macroeconomists are to be believed) at some point decided inflation was coming.

ds: Not sure if you're referring to me or Nick Rowe. *I* think it is entirely fair to say that macro-models use forward-shifting consumption to explain how inflation expectations leads to actual inflation. Maybe I'm wrong, but if Nick's taken me to task for this I've missed it.

A typical inter-temporal consumption function in micro is:


c1+(1+i)/(1+r)*c2=m1+m2(1+r)

i=inflation as I don't know how to make pi

So your decision to consume today or tomorrow (or conversely, save more or less) depends on your discount rate (combo r and i).

Nick's submission explicitly moved consumption forward (stocking up on scotch!), as he bought a number of luxury items plus some home repair (as well as other things).

NICK: If you can explicitly call out the portion of the DSGE model to show how inflation expectation impacts inter-temporal consumption decisions, I'd be obliged. But I'm guessing the basic micro equation I pulled out isn't too far removed.

9:06 AM  
OpenID pebird said...

Nick:

"Your answer will be graded by: an accountant; a businessman; a Russian day-trader; and an economics professor."

If that is the case, why does the student have to write clearly?

8:00 AM  
Blogger Nick Rowe said...

Sorry for the delay.

Short Market:

I don't understand your criticism.

When we are asked "what is the effect of an increase in expected inflation?", it's normal to reply by assuming the system is intitially in equilibrium, then assuming an exogenous shock hits (e.g. people's expectations of inflation suddenly increase), and then explain how that exogenous shock moves the economy to a new equilibrium.

For example, "What's the effect of pressing down on the gas pedal?" Answer: "assume you are currently driving at a steady 50km/hr, then suddenly you press down on the gas pedal and hold it down, then your car will accelerate and eventually reach a speed of 100km/hr".

"Everybody knows that laptop prices will be lower tomorrow but people still buy them today. Everybody knows that oil prices will be higher tomorrow yet oil producers still sell oil today."

Of course. But they buy *fewer* laptops today than they would if they thought laptops would *not* be cheaper tomorrow. And they would buy *less* gas today if they didn't think gas prices would be higher tomorrow.

Greg: "So Nick
What is the tipping point? How many people with inflation expectations does it take to drive prices up right now?"

Just one. But one person's increased demand will only have a miniscule effect, of course. Very roughly, if 1% of the population expect hyperinflation, and the other 99% expect none, the effects will probably be roughly 1% the size of the effects if everybody expected hyperinflation. That would only work exactly if all demand and supply functions etc. were linear of course, and it were a random sample of the population.

"It seems to me from your story that if we start hearing about people walking into an Athens computer store with $1,000 dollars wishing to purchase a laptop and they are denied, not because they dont have one for 1000$ but because they want to sell it for more next month, then we should start to worry. Has this behavior EVER been observed?"

No. That's not my story at all. My story is that they *won't* be denied. My story is that they *will* be allowed to buy a laptop, but at a higher price. Has this ever been observed (an increase in demand and/or reduction in supply causing an increase in price)? Yes.

4:53 AM  
Blogger Nick Rowe said...

Nolan: substitution vs income effects. Good question! And yes,

"It's true that I only took a handful of economics courses as an undergraduate, but I must say I never really heard a good explanation from any of my professors of why substitution effects should dominate income effects in certain markets or why the opposite should be true in other markets. Perhaps this is addressed more fully in graduate school econ classes."

Because very few profs ever do give a good answer to that question. And that's probably because very few profs actually understand the answer themselves. That is a definite failing of economics education, in many cases.

Here's my answer to your excellent question:

For every apple bought there's an apple sold. An increase in the price of apples reduces the real income of apple buyers, and increases the real income of apple sellers by an exactly equal amount. At the macroeconomic level, there is no *aggregate* change in real income. What there is is a change in the *distribution* of income between apple sellers and apple buyers. So the income effects go in opposite directions for buyers and sellers.

Will the income effects exactly cancel out? No. Or only under very special assumptions about preferences (for example, identical homothetic utility functions). But this does mean that at the aggregate level the income effects could go in either direction. And you would need very extreme assumptions to get the income effects to be bigger than the substitution effect.

There is only one empirically validated example of a Giffen good that I'm aware of. Rice, in some poor village in China. (And I'm not sure that even here they have captured the macro-effects properly).

And in the case of real interest rates on the savings/consumption choice: remember that for every $1 borrowed there is $1 lent. Just like apples.

And yes, teaching economics at first year university is hard. And it must be harder still in high schools. And even harder to explain it quickly to school board members. I definitely hear you there.

Viviane: "My mum (and all family) used to buy dollars."

Another good answer. I forgot that one. And that causes the exchange rate to depreciate, and has an effect on the CPI via the price of imports.

5:14 AM  
Blogger Nick Rowe said...

Winterspeak:

Sorry I should have replied to this waaay earlier.

"A typical inter-temporal consumption function in micro is:

c1+(1+i)/(1+r)*c2=m1+m2(1+r)"

Basically yes. But there's one big difference when it comes to the general equilibrium experiment in macro. It's that m2. I take it that's (expected) money income? That will be affected by (expected) inflation.

Let y1 and y2 be real income, this year and next. And c1 and c2 be real consumption. And n be real interest rate (nominal minus pi). (I can't do pi either).

Then the intertemporal budget constraint is

c1 +c2/(1+n) = y1 + y2/(1+n).

That works at the macro level, where y is GDP (or, properly, NNP). At the micro level, we have to add credits for creditors, and subtract debts for debtors.

5:27 AM  
Blogger Nick Rowe said...

Winterspeak: "NICK: If you can explicitly call out the portion of the DSGE model to show how inflation expectation impacts inter-temporal consumption decisions, I'd be obliged. But I'm guessing the basic micro equation I pulled out isn't too far removed."

It's the Euler equation.

Roughly,

MU(c1)/MU(c2) = 1/(1+r-pi)

where r is the nominal interest rate, pi is expected inflation, so n=r-pi, and MU is marginal Utility, which depends negatively on consumption. (Add in a constant for time preference if you want).

So when you maximise U(C1) + U(C2) subject to the budget constraint, the above is what you get.

5:37 AM  
Blogger winterspeak said...

Nick: Thanks! And yes, inflation is part of the discount rate and it's modeled as an inter temporal consumption decision. I think this is the wrong way to model this element of the economy.

Also, your "every dollar lent=every dollar borrowed" is why conventional economics does not understand the macro economy. The dollars lent are brand new dollars -- this is not a redistribution (whereas every dollar spent is a dollar earned *is*). The problems worse at the Federal level.

7:38 AM  
Blogger Nick Rowe said...

Winterspeak: aha! I will respond on the other post.

9:57 AM  

Post a Comment

Subscribe to Post Comments [Atom]

Links to this post:

Create a Link

<< Home