The distinction you have made in your last posts between the difference between movements in relative prices and prices in general is crucial for a proper understanding of inflationary factors and forces.
You also list a variety of reasons behind changes in relative prices often confused for general inflationary price changes.
But it is worth pointing out there is a connection between monetary changes that end up influencing prices in general and relative prices. This factor has been emphasized since Richard Cantillon and David Hume in the 1700s, John E. Cairnes in the 1800s, and “Austrian” economists like Ludwig von Mises and F.A. Hayek in the 1900s.
All changes in the supply of money are injected into (or withdrawn out of) the market somewhere and proceeds to pass from hand to hand as changes in the demands for various goods in some temporal sequence. The structure of relative prices and wages, profit opportunities and resource uses (labor, capital), are effected for as long as the monetary changes continue and have no fully and finally had there impact throughout the economy.
Thus, changes in prices in general (price inflation or price deflation) are never completely separable from relative price changes due to the inherent “non-neutrality” of money and monetary changes.
I’ve been reading your most recent posts with great interest. Economic insight is especially welcome since it allows for better understanding.
In the midst of reading this I recall a recent essay on Veblen ‘goods’ versus Giffen ‘goods’. My question would be are these a one off or do you feel they have a discernible effect on economic outcomes?
"This is profoundly disconcerting. I expected that price levels would be cointegrated, and converge in the long run. It might take a while, but they should converge eventually."
I had exactly the same idea. So I did the time series analysis for the aggregated HICP-indices of the 12 founder states of the European Monetary Union from from January 1999 to September 2019. The results for the 66 country pairs in short: Price levels follow different random walks with different drift parameters; the resulting real exchange rates between these countries are not stationary but random walks too; relaxing assumptions and testing for cointegrating relationships only shows that there are no systematic cointegration relationships either.
After facing these results, I had again exactly the same idea as you "I presume this pattern comes down to “nontradeables,” things like the price of housing and services like haircuts."
So, I applied the same empirical research to 8 subcomponents of the HICP: 4 tradables and 4 non-tradables. The findings show that there is no systematic difference between the behavior of the HICP components for tradable goods and non-tradable services. Out of the 594 potential cointegration relationships examined, 112 fail to reject the null hypothesis that a cointegration vector exists at the conventional significance level of 5%. However, in a mere 40 cases, the estimated cointegration parameters display the theoretically expected signs. The number of cointegration vectors for “total goods” matches that of “total services” despite involving different pairs of countries. It is not possible to identify specific country clusters where the existence of cointegrating relationships is less frequently rejected than in other cases. Furthermore, there is no discernible clustering around northern or southern European countries. (Would of course be interesting to have a similar empirical analysis for the US Federal States...)
To me these results are indeed disconcerting! What makes the situation not much better is the fact that the European Central Bank does not want to discuss these problems. When I submitted my paper first to the International Journal of Central Banking the editor rejected my paper without a referee process. The editor was Luc Laeven, the Director-General of the Research Department of the European Central Bank.
I guess the ECB has more important things to do - like the "Greening of Monetary Policy" ;-)
Perhaps the answer lies in breaking down the index to try to see which of the products in the basket are putting pressure on this difference in price levels.
This lays out the difference between relative price changes and inflation in a way that makes perfect sense. A “shock” that changes supply or demand for a good can change the relative price of that good, it but cannot change average price of all goods without monetary policy action. The purpose of monetary policy -- indeed the purpose of money -- is to facilitate adjustment in relative prices so as to maximize real income -- to make Say's law true in practice.
The rationale Cochrane lays out leads directly to Flexible Average Inflation Targeting (FAIT) or [or Flexible Nominal GDP Level Targeting (FNGDPLT)] as a monetary policy regime. The average level of shocks -- the Brownian movement of preferences and technology in a dynamic economy -- combined with the average amount of price stickiness leads to a central bank setting a non-zero target for average inflation. (A target is forward looking and so should be the "A" of FAIT) When extraordinary shocks come along the central bank temporarily increases the rate of inflation to facilitate the extraordinary relative price adjustments needed to preserve market clearance, hence Flexible Average Inflation Targeting; there will be temporary periods of rising over-target inflation followed by periods of restraint when the central bank brings inflation back down to target.
The exact nature of the shock is not important, only its disparate impact on prices that requires relative price adjustment. An aggregate demand or supply shock that is approximately uniformly distributed across sectors (hypothetically possible) would not occasion over-target inflation at all, but only changes in the interest rate (or whatever vector of monetary policy instruments being used).
In practice any extraordinary shock would call for some over-target inflation. This would give the impression of the shock, the increase in the deficit, or tariffs (even if deficit neutral) for example, "causing" the over-target inflation.
If I had to criticize anything, it would be that, reading between the lines, I conclude Cochrane thinks that the FIAT should be zero, and possibly with no “F.”
The Katrina supply shock distorted the CPI data from 2004-2008. CPI hit 5.5% year over year in July of 2008…buh Housing Bubble. The inflation of 2004-2008 is much more interesting than this round of inflation which is similar to post WW2 inflation from bullwhip effect.
I'd like to see a post explicitly pointing out that there's no such thing as "housing inflation" or "health care inflation" or "meatball inflation." E.g. housing prices can be written as the sum of the (common!) inflationary component plus its idiosyncratic component, and that the overall change can imply something is more or less affordable, but that the overall change should not be referred to as inflation. The press have gone nuts with phrases like "housing inflation" in the last few years and I've even seen economists start using it, perhaps because they think it's easier to reach people with that bogus terminology.
"I presume this pattern comes down to “nontradeables,” things like the price of housing and services like haircuts."
This sounds fair. I would also add the prices of energy which are very much dependent on the tax system (gasoline at the pump is way cheaper in Luxembourg than in France for instance) but also on the sources of energy (Germany was highly dependent on Russian gas because of the pipelines built by the Comecon while France was not).
The distinction you have made in your last posts between the difference between movements in relative prices and prices in general is crucial for a proper understanding of inflationary factors and forces.
You also list a variety of reasons behind changes in relative prices often confused for general inflationary price changes.
But it is worth pointing out there is a connection between monetary changes that end up influencing prices in general and relative prices. This factor has been emphasized since Richard Cantillon and David Hume in the 1700s, John E. Cairnes in the 1800s, and “Austrian” economists like Ludwig von Mises and F.A. Hayek in the 1900s.
All changes in the supply of money are injected into (or withdrawn out of) the market somewhere and proceeds to pass from hand to hand as changes in the demands for various goods in some temporal sequence. The structure of relative prices and wages, profit opportunities and resource uses (labor, capital), are effected for as long as the monetary changes continue and have no fully and finally had there impact throughout the economy.
Thus, changes in prices in general (price inflation or price deflation) are never completely separable from relative price changes due to the inherent “non-neutrality” of money and monetary changes.
I’ve been reading your most recent posts with great interest. Economic insight is especially welcome since it allows for better understanding.
In the midst of reading this I recall a recent essay on Veblen ‘goods’ versus Giffen ‘goods’. My question would be are these a one off or do you feel they have a discernible effect on economic outcomes?
"This is profoundly disconcerting. I expected that price levels would be cointegrated, and converge in the long run. It might take a while, but they should converge eventually."
I had exactly the same idea. So I did the time series analysis for the aggregated HICP-indices of the 12 founder states of the European Monetary Union from from January 1999 to September 2019. The results for the 66 country pairs in short: Price levels follow different random walks with different drift parameters; the resulting real exchange rates between these countries are not stationary but random walks too; relaxing assumptions and testing for cointegrating relationships only shows that there are no systematic cointegration relationships either.
After facing these results, I had again exactly the same idea as you "I presume this pattern comes down to “nontradeables,” things like the price of housing and services like haircuts."
So, I applied the same empirical research to 8 subcomponents of the HICP: 4 tradables and 4 non-tradables. The findings show that there is no systematic difference between the behavior of the HICP components for tradable goods and non-tradable services. Out of the 594 potential cointegration relationships examined, 112 fail to reject the null hypothesis that a cointegration vector exists at the conventional significance level of 5%. However, in a mere 40 cases, the estimated cointegration parameters display the theoretically expected signs. The number of cointegration vectors for “total goods” matches that of “total services” despite involving different pairs of countries. It is not possible to identify specific country clusters where the existence of cointegrating relationships is less frequently rejected than in other cases. Furthermore, there is no discernible clustering around northern or southern European countries. (Would of course be interesting to have a similar empirical analysis for the US Federal States...)
The detailed results are published in the Journal of International Financial Markets, Institutions and Money: https://www.sciencedirect.com/science/article/abs/pii/S1042443122001263
An open access overview article is available under: https://www.intereconomics.eu/contents/year/2023/number/6/article/the-divergence-of-price-levels-in-the-european-union.html
To me these results are indeed disconcerting! What makes the situation not much better is the fact that the European Central Bank does not want to discuss these problems. When I submitted my paper first to the International Journal of Central Banking the editor rejected my paper without a referee process. The editor was Luc Laeven, the Director-General of the Research Department of the European Central Bank.
I guess the ECB has more important things to do - like the "Greening of Monetary Policy" ;-)
I'm puzzled. 🤔
Perhaps the answer lies in breaking down the index to try to see which of the products in the basket are putting pressure on this difference in price levels.
I'll try to see if it's possible to decompose the index this way.
This lays out the difference between relative price changes and inflation in a way that makes perfect sense. A “shock” that changes supply or demand for a good can change the relative price of that good, it but cannot change average price of all goods without monetary policy action. The purpose of monetary policy -- indeed the purpose of money -- is to facilitate adjustment in relative prices so as to maximize real income -- to make Say's law true in practice.
The rationale Cochrane lays out leads directly to Flexible Average Inflation Targeting (FAIT) or [or Flexible Nominal GDP Level Targeting (FNGDPLT)] as a monetary policy regime. The average level of shocks -- the Brownian movement of preferences and technology in a dynamic economy -- combined with the average amount of price stickiness leads to a central bank setting a non-zero target for average inflation. (A target is forward looking and so should be the "A" of FAIT) When extraordinary shocks come along the central bank temporarily increases the rate of inflation to facilitate the extraordinary relative price adjustments needed to preserve market clearance, hence Flexible Average Inflation Targeting; there will be temporary periods of rising over-target inflation followed by periods of restraint when the central bank brings inflation back down to target.
The exact nature of the shock is not important, only its disparate impact on prices that requires relative price adjustment. An aggregate demand or supply shock that is approximately uniformly distributed across sectors (hypothetically possible) would not occasion over-target inflation at all, but only changes in the interest rate (or whatever vector of monetary policy instruments being used).
In practice any extraordinary shock would call for some over-target inflation. This would give the impression of the shock, the increase in the deficit, or tariffs (even if deficit neutral) for example, "causing" the over-target inflation.
If I had to criticize anything, it would be that, reading between the lines, I conclude Cochrane thinks that the FIAT should be zero, and possibly with no “F.”
The Katrina supply shock distorted the CPI data from 2004-2008. CPI hit 5.5% year over year in July of 2008…buh Housing Bubble. The inflation of 2004-2008 is much more interesting than this round of inflation which is similar to post WW2 inflation from bullwhip effect.
I'd like to see a post explicitly pointing out that there's no such thing as "housing inflation" or "health care inflation" or "meatball inflation." E.g. housing prices can be written as the sum of the (common!) inflationary component plus its idiosyncratic component, and that the overall change can imply something is more or less affordable, but that the overall change should not be referred to as inflation. The press have gone nuts with phrases like "housing inflation" in the last few years and I've even seen economists start using it, perhaps because they think it's easier to reach people with that bogus terminology.
"I presume this pattern comes down to “nontradeables,” things like the price of housing and services like haircuts."
This sounds fair. I would also add the prices of energy which are very much dependent on the tax system (gasoline at the pump is way cheaper in Luxembourg than in France for instance) but also on the sources of energy (Germany was highly dependent on Russian gas because of the pipelines built by the Comecon while France was not).
OT but begging for a reply.
Yes, Bank Indonesia (BI) has been buying government bonds as part of its debt monetization program:
2021: BI bought about 215 trillion rupiah of government bonds
2022: BI bought 224 trillion rupiah of government bonds
2024: BI became the largest holder of Indonesia's sovereign debt, with 23% of the total rupiah bonds
BI's ownership of government bonds allows it to:
Stabilize bond yields to prevent outflows during market volatility
Curb volatility during an unfavorable global market environment."
---30---
Indonesia now has below-target inflation, sub 2%.
Of course, Japan did the same thing forever.
Is this the solution to the US debt bill?
We can rest assured Congress and Trump are not going to balance the budget.
Yet central banks that build balance sheetห do not result in inflation. But they do remove debt from taxpayer shoulders.