The Money Printing Era Will Implode In One Or Two Years

[Marginal_Interest]

David,

You certainly have.

I think I see where we’re differing here. I’m looking at measuring inflation once an injection of money has worked its way through the system where it will eventually raise all prices by the same proportion. Ie, the injection of money will only end up creating an income effect. It won’t distort relative prices, and so any relative price distortion that occurs because of non-monetary factors (supply shocks) should be removed from the measure, because they aren’t representative of underlying inflation.

If I understand correctly, you’re saying that in the short run an injection of money won’t raise all prices in the same proportion. There will be some relative price distortions until the money works its way through the system. In which case, removing the substitution effects will be removing price rises that the increase in the money supply has caused, which means that the index will under-estimate the effects of a monetary injection.

And that’s completely true. However, I’m not saying that making weekly or monthly adjustments for substitution is going to accurately measure inflation. I’m saying that adjustments to at least quarterly, but especially 6 monthly to yearly measures will have allowed enough time to pass for all prices to be raised in proportion. So compensating for substitution in these longer periods won’t create a downward bias.

Now on your supply side price change point, that doesn’t constitute inflation pecisely because it’s a change in relative price due to supply shocks, not the general price due to an increase in the money supply. We only want to compensate people for changes in prices that occur due to monetary inflation. If there’s a cyclone that destroys a large part of a banana crop, the price of bananas will skyrocket. The relative price has been changed and people will naturally begin substituting apples for bananas. We don’t compensate a retiree for the increase in the price of bananas, despite that bananas are in the basket of goods, because everybody is rationally substituting away from them because of real factors. We only want to compensate them for price increases that are the result of the increased money supply, not for real shocks.

Headline CPI is not adjusted for the substitution effect by the way (I probably should have said that earlier). Chained CPI is. So the monthly CPI measure you see in the news doesn’t do this stuff. Social Security isn’t actually adjusted with chained CPI, it’s adjusted with CPI-W, which is a wage index (so increases in the general level of wages means that retirees get an increase in social security payments). Now TIPS, the thing that started this conversation, are index to CPI-U (headline CPI). Since coupons are paid every 6 months (which is a reasonable amount of time for the chained CPI to be accurate), they’re actually getting paid slightly more than “real” inflation since the measure also includes supply shocks, not just the inflation caused by the increase in money.

[djsherin]

Marginal Interest,

This is one of the more interesting conversations I’ve had in a while and I’m enjoying it.

I think you adequately pointed out what the difference in thinking has been, but I’m not a subscriber to the neautrality of money in the long run. Essentially I don’t see any reason to say apodictically that all income must rise by the same proportion from an injection of money. Certainly those on fixed (non-adjustable) incomes don’t see a rise, but more fundamentally, since someone gets the new money first (in my example, the buyers of x), the industry where that money is spent will have the chance to expand, conduct R&D, invest, etc. more than it otherwise would have and before any other firm has. This could easily lead to a different pattern of production in the long run with the initial injection of money than the long run without the initial injection of money. I guess this isn’t really the place to debate this particular topic, but I did want to point it out.

That said, I would agree that the long run with the injection of money will be closer to our initial situation before the money supply increased than the short run would be. Depending on how different the long run and the initial are from one another would determine how realistic the assumption of long run money neutrality is. Of course this is impossible to test since we can’t hold all other factors constant.

But let’s say all income does rise by the same proportion and relative prices don’t change (they all rise by the same proportion), then you wouldn’t need to compensate for substitution because the pattern of consumption stays the same (give constant preferences), i.e. because there is no substitution. Just to illustrate (and I’m not saying you disagree, I’m just being as thorough as possible), if income rises by A% and all prices rise by A%, then the consumer’s budget constraint doesn’t change and since we’re assuming constant preferences, the consumer would attain the same point on his indifference curve. I’m not a huge fan of indifference curve analysis, but I want to make sure we’re on the same page. The only change in this case would be a higher price level.

On supply, I’m not denying that it is a relative price change, but I don’t see how it isn’t also a general price change. If half the prices in CPI go up by B% and half the CPI go down by B%, you have a relative but not general change in prices. However, in my example, two prices go up while others presumably remain constant which would lead to an increase in the CPI (probably a small one, but an increase nonetheless). Shown with an aggregate supply and demand curve, the aggregate supply curve would shift a tad to the left leading to an increased price level and a lower level of GDP.

I’m not specifically talking about compensating retirees and how much they should be given, but rather what constitutes a rise in the general price level. Perhaps retirees ought not to receive more compensation from a decreased supply, but the general price level is still higher and thus the cost of living is likewise higher.

I’m aware of the differences in the CPI calculations. I guess I’m approaching this from more of a theoretical, and less of an applied, angle so I didn’t think it mattered much to say which CPI we were talking about. I see your point about the CPI-U vs. Chained CPI, but I think this goes back to whether money is neutral in the long run (if it is, your point is valid; if not I would say it isn’t), and how long it takes to reach the long run (if 6 months is a reasonable amount of time for the long run to be attained, your point is valid; if not, again, I don’t think it would be). But anyway, I have to get going so I’ll end my response here. Don’t take anything I say personally btw, it’s been a good discussion.

[Marginal_Interest]

@David Sherin

This is one of the more interesting conversations I’ve had in a while and I’m enjoying it.

Me too. It’s one of the more reasonable conversations I’ve had with somebody of the Austrian persuasion.

I think you adequately pointed out what the difference in thinking has been, but I’m not a subscriber to the neautrality of money in the long run. Essentially I don’t see any reason to say apodictically that all income must rise by the same proportion from an injection of money. Certainly those on fixed (non-adjustable) incomes don’t see a rise, but more fundamentally, since someone gets the new money first (in my example, the buyers of x), the industry where that money is spent will have the chance to expand, conduct R&D, invest, etc. more than it otherwise would have and before any other firm has. This could easily lead to a different pattern of production in the long run

If I said “income rises in the same proportion”, that was a mistake. Sorry. I meant to say “wages”. As you point out those on fixed incomes don’t only experience a rise in prices.

I’m not sure about the “some industries get the money first” argument. I see no reason why an increase in bank reserves results in only lending to certain industries first. Everybody will have access to the credit and it’ll go to where it’ll get the highest return. Nobody has “dibs” on it. And let’s not forget that we don’t need the new money to reach certain people before they can spend it to raise prices. Those servicing debt on adjustable rates see an effective increase in income through not having to repay so much each month.

As for long run neutrality of money, I think there’s considerable evidence to that.

That said, I would agree that the long run with the injection of money will be closer to our initial situation before the money supply increased than the short run would be. Depending on how different the long run and the initial are from one another would determine how realistic the assumption of long run money neutrality is. Of course this is impossible to test since we can’t hold all other factors constant.

Not being able to hold all factors constant doesn’t mean it’s impossible to test. Even if you can, would you be saying “well yeah you can hold all factors constant, but you can’t know which distribution the variable comes from and you can’t perform infinity trials”. We come up with creative new ways to deal with problems like that. For the “not knowing which distribution”, we use the central limit theorem; for not being able to run infinity trials we have confidence intervals and minimum sample sizes for the CLT to hold reasonably well. If we end up with a variable that’s correlated with the error term, we come up with the method of instrumental variables. Limitations like that don’t mean that you can’t draw any conclusions.

On supply, I’m not denying that it is a relative price change, but I don’t see how it isn’t also a general price change.

I guess it depends on how we’re defining “general”. I’m saying “all prices increase in the same proportion” rather than “an increase in the price level aggregate”. Supply shocks do increase the price level, but we don’t technically call it inflation. Again it depends strongly on your definition of inflation. Some people go with “an increase in the CPI”. I’m using a definition in the spirit of “inflation is everywhere and always a monetary phenomenon”.

I’m aware of the differences in the CPI calculations. I guess I’m approaching this from more of a theoretical, and less of an applied, angle so I didn’t think it mattered much to say which CPI we were talking about. I see your point about the CPI-U vs. Chained CPI, but I think this goes back to whether money is neutral in the long run (if it is, your point is valid; if not I would say it isn’t), and how long it takes to reach the long run (if 6 months is a reasonable amount of time for the long run to be attained, your point is valid; if not, again, I don’t think it would be). But anyway, I have to get going so I’ll end my response here. Don’t take anything I say personally btw, it’s been a good discussion.

Well in that case I think it comes down to: if money is neutral in the long run (which I accept), CPI overstates inflation and chained CPI is more accurate; in which case compensation (say for TIPS) based on headline CPI is actually over-compensation.If money is non-neutral in the long run, CPI does not overstate inflation; but there’s also no reason it would understate it.

[djsherin]

I haven’t forgotten about this, I’ve just been really busy and only had sporadic internet access. I’ll try to respond as soon as I can but it may be a few more days.