Inflation is higher than you think

The Covid-19 Pandemic has led to changes in consumer expenditure patterns that can introduce significant bias in the measurement of inflation. I use data collected from credit and debit transactions in the US to update the official basket weights and estimate the impact on the Consumer Price Index (CPI). I find that the Covid inflation rate is higher than the official CPI in the US, for both headline and core indices. I also find similar results with Covid baskets in 10 out of 16 additional countries. The difference is significant and growing over time, as social-distancing rules and behaviors are making consumers spend relatively more on food and other categories with rising inflation, and relatively less on transportation and other categories experiencing significant deflation.

That is from Alberto Cavallo, and as for concrete numbers: “The Covid Core deflation in April was only half of that in the Core CPI, while theannual inflation rate is at 1.73% compared to the 1.43% in the official Core index.”  And that is not accounting for the disappearing goods bias: “For example, the share of products with missing prices in the US CPI rose from 14% in April 2019 to 34% in April 2020.”

Of course this also has implications for those insisting we should think of this primarily as a demand shock.

Comments

Actually, it doesn't have implications for the demand shock claim. For evidence of demand shocks you want to look at NGDP growth, not inflation. And NGDP growth has done far worse than inflation.

Respond

Add Comment

Price increases aren't a measure of inflation. But enpixelating trillions of dollars in an effort to salvage the lock-down disaster is. That's why the stock market is booming and real estate prices continue to climb. Savvy folks with cash know it's just a matter of time before those phony dollars won't buy near as much as they once did. They're putting them in stocks for the purpose of liquidity and in property because it's a hedge against inflation. Putting money under your mattress is almost the same as setting it on fire.

"Price increases aren't a measure of inflation"
seriously?

Key is the word "measure." Also key is your definition of inflation. If you define inflation as price increases, you're right. If you define it as increases in the money supply, you're wrong, particularly in MEASURING inflation. The expression "price inflation" conflates the two approaches; it is terminologically destructive.

gracias
is the consumer price index a measure of inflation?

measure- def. estimate or assess the extent, quality, value, or effect of (something).

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Then why are they also putting a ton of money in bonds? I don't think the market is by any means forecasting high inflation right now.

Respond

Add Comment

"Price increases aren't a measure of inflation."
Ohhh ok.
"Savvy folks with cash know it's just a matter of time before those phony dollars won't buy near as much as they once did. "

So price increases are inflation after all. But I guess this is trying to say the inflation will arrive in the future but it may tarry. But then what is it doing right now if it can't be measured in prices?

Ludwig von Mises: If you increase the quantity of money, you bring about the lowering of the purchasing power of the monetary unit.

Yes which means higher prices. That assertion is disproven if money quantity is increased but I do not see higher prices. If gov't prints more $100 bills but the McDonald's dollar menu stays $1, the purchasing power of the monetary unit has not been decreased.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

A negative demand shock alone would lower the price level. A negative supply shock alone would raise the price level.

I doubt it. A default raises the price level, an undefault lowers the price level. One actually has to unilaterally remove or add to the loose money.

Wait until the federal defaults and you will see price level rise. Otherwise we simply alter our budgets, removing or adding categories. We utilize the double entry accounting system which is based in preserving price level.

Respond

Add Comment

Respond

Add Comment

OMG

That means
There will be
More
Commercials for
Gold on Fox TV.

Respond

Add Comment

The result will always be a symmetric bell shaped curve, by assumption in the technique. Like Scott says.
Call it what you like, but it will always be true as an equilibrium. Stores quit selling things at a loss and everything sold at equilibrium will always appear as a bell shaped curve because the equilibrium condition of specifies nearly independent arrivals.

Respond

Add Comment

No. Inflation is higher than economists thought. Regular folks that go grocery shopping already knew this.

Economists should be careful to tell us average punters whether they are talking about monetary inflation, or changes to purchasing habits, or prices (when the other two don't change).

For what it's worth, my purchasing habits are tremendously changed. I used to be a fairly price-sensitive shopper. I'd refuse "high" prices, and pick up an item later, elsewhere.

Now I flat don't care, and I'll load my basket fully on a Costco or Trader Joes run. Obviously the market understands that, and prices accordingly.

And the supermarkets in my city, though they still send those weekly flyers that advertise their sale prices and coupons, are offering fewer such price deals and with smaller discounts than before.

The flyers are still the same size, but they increased the size of the images. There are fewer discounts being offered by them now.

I've forgotten the answer to this Econ 101 question: does the BLS account for sales and coupons when it collects market data? IIRC there was an article 30 or 40 years ago that said that such discounts are important.

Sales, yes. Coupons, no.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

I think you're implying the disappearing goods bias increases inflation. I can't buy a NBA ticket. They don't want to sell it because they think I don't want to buy it. Perhaps they are right. This means the price decreased. In this case disappearing goods decreases inflation.

Respond

Add Comment

The inflation rate has to go up in the months ahead as the economy re-opens and people start buying goods again. It has to: producers can't just flip a switch and start producing again. And then there are the broken supply chains. Except in war time I don't believe we have ever had a combination supply shock and demand shock like this. As for Sumner's comment, he seems content viewing the crisis through the lens of historical shocks, whether supply shocks or demand shocks. This is different.

I know: the shortage of supply may well be offset by the shortage of demand. Looking backwards the past few months is misleading: income actually went up, but it's an illusion, attributable as it is to government money drops. Republicans claim they won't support any more money drops. Claim. If the economy starts to sink in late summer, the Republicans will show us how to swim.

Respond

Add Comment

Respond

Add Comment

Wow. Alberto has changed the weights of goods and services to estimate a new CPI and he founds that the new CPI indicates an increased level over the past few months. I'm sure that Alberto's father (also the father of Argentina's hyperinflation in 1982) and all other Argentinian economists are not surprised by Alberto's finding. In Argentina, more than 50 years ago, Julio Olivera used to explain high inflation as a result of large changes in relative prices and an accommodating money supply, and for that, he used examples of alternative CPIs --same goods and services but different weights. Indeed, under Milton Friedman's definition, it is not inflation.

Respond

Add Comment

It's difficult for me to think of a less useful and less coherent variable in all of economics than CPI. Maybe of those those goofball Saez-Zucman wealth inequality measurements?

Respond

Add Comment

It’s both a demand shock and a supply shock I think but the sectors where each happens may have limited overlap.
It’s a demand shock for airline travels and hotels and Uber and restaurants and gasoline for example even after the reopening.
It’s a supply shock for some grocery items and for my Nintendo Switch ( I am still waiting for supply to increase and the price to increase.
I have also noticed that some notebook computers have long lead times.
What’s the overall impact on inflation , I am not sure.

Respond

Add Comment

* and the price to decrease) It’s ~ $500 right now. It used to be around $300.

Respond

Add Comment

The difference is significant and growing over time, as social-distancing rules and behaviors are making consumers spend relatively more on food and other categories with rising inflation, and relatively less on transportation and other categories experiencing significant deflation.

A couple of big problems here. First, shifting spending from other categories to food means Americans are spending much less overall (even slightly inflated food is cheap). And there has also been a big shift from restaurant to home-cooked meals which, again, means a big cost savings.

Indeed, 'relative' is hiding a lot here. Spending less on a two week vacation overseas and more on groceries is not an even swap by any means. Inflation doesn't make sense if I'm meeting my daily needs with less swipes of my debit card than I did before.

Respond

Add Comment

Respond

Add Comment

Question. Let's say I used to buy 100 things for $1 each and for years prices never change so zero inflation.

Now because of shutdown I've stopped buying 50 of those things. My demand reduction has caused their price to drop $0.50. I've started buying more groceries, though, so let's say that's 25 things that have gone from $1 to $2 in price.

Using the old basket of goods:
50 things @ $0.50 = $25
25 things @ $1.00 = $25
25 things @ $2.00 = $50

Total $100 / 100 things = 1.00 no inflation.

Now, though, say I construct a new basket:
25 things $1 = $25
25 things $2 = $50

Total = $75 divided by 50 things = 1.50 where as I used to be able to buy 75 things for $75 so using the new basket I declare 50% raging inflation.

Yet this doesn't seem quite right. I used to spend $100 to get my needs and now I only spend $75 yet 50% inflation? But isn't part of this mess coming from entirely recomputing the basket based on a one time event? Are we, for example, going to tear up airports to plant vegetables? I mean if you really think people will never fly again for a generation or two because of this then that would make sense. And yet the stock market does not look at Boeing the way it looked at horse driven buggy makers after Ford's revolution.

That's the classic price index problem. It's usually covered in intermediate micro classes (rather than macro, because it's useful to use budget constraints and indifference curves to illustrate what's going on with the old vs new basket).

There is no ideal solution. If you stick with the old basket, you're using what's called a LaSpeyres price index. It will generally overstate what the true inflation rate is.

If you use the new basket you're calculating a Paasche price index. It will generally understate the true inflation rate.

LaSpeyres price indices need less updating but obviously you need to update the contents of the basket of goods occasionally. I believe the BLS updates the basket every two years, although some years ago it also started reporting a chain-weighted CPI which is updated monthly.

What about new goods that didn't exist at all in the old basket?
Needless to say they're going to be poorly measured, and most price indices will overstate inflation to the extent that they fail to account for new goods; Alex and Tyler like to point this out periodically.

It's rarer, but the opposite effect occurs when a good is no longer available. E.g. if you liked to keep pet passenger pigeons, or if you wanted to read Plutarch's "Life of Scipio Africanus", you're out of luck compared to having the old basket that contained those goods.

And as Alex and Tyler also like to mention, what about improvements in quality; what kind of satisfaction do you get from buying a compact car today compared to one 30 years ago? In theory a correctly measured basket should properly account for the quality changes but in reality that's exceptionally hard to do.

So here we don't really have new goods or old goods no longer available but simply a decision, presumably temporary, to stop consuming some goods.

Some of these goods are gone forever. The haircuts I did not get in April, May and June will not result in me getting 4 haircuts in July. Many goods, though, are simply stockpiled. The oil that went into storage (or just wasn't pumped) when prices went negative, can be burned later on. So can a lot of things.

We have a demand shock and a supply shock but its a positive supply shock rather than a negative one.

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment

Rent or its mortgage + tax equivalent is the single biggest expense, and I've only gotten anecdotal data on it. Are all that many people actually moving or negotiating to get lower rents? Right now, enhanced unemployment insurance is hiding any real effect. When it runs out, we'll see what happens.

In the US, people started to stay in, cut back on travel and unnecessary shopping before any official shutdowns, so one could argue that there was a demand side shock. Once the shutdowns began, this was augmented by a supply side shock. The recovery is going to be shaped by people's fears and expectations. You can open the venues, but you can't force people to attend.

My guess is that it is the better off people, the ones with more money to spend and who are generally more in control of their lives and health, who are going to be hardest to persuade to resume spending.

I'm sorry, oil went negative. Someone tell me how that indicates a supply side shock.

Respond

Add Comment

Respond

Add Comment

The bean counters have an index of consumer goods, trimmed PCE. Trimmed means make it Gaussian. It is an assembly of negative price changes and positive price changes. It has a flaw, zero, suppliers exit the business. We are confused, we think the reconciliation is done. It isn't. Price revision go back. If there was a general price rise, absent federal default, then the accounting system is expensing it. What we think is general inflation is really price variance under sampled. Bankers should not be doing price indices. They will always get caught in a correction, as pricing is a dual asymmetric process.

Better idea.
Get the consumer list of prices in and supplier list of prices out.
Threat them as independent sequences that must be matched, by moment .

Respond

Add Comment

Respond

Add Comment

Amazing that the higher inflation rate is still under the long term 2 percent that is supposed to have been the target for the last few decades.

Respond

Add Comment

There were basic items which competitive markets would be on sale simply always. Not at the same store, though. It would rotate. But in urban areas with lots of options, there would always be one chain grocery store with pork tenderloin, or ground beef, or pasta, or diapers, or eggs, or frozen pizzas, whatever... items which are staples for many and would always be available as a "loss leader" somewhere. Those sales are gone.
Take out places have also axed all their special weekday deals. There's not "Two for One Tuesday's" at pizza places anymore, traditionally the slowest day of the week so they offered a discount to get you to buy more and make the same profit.

Respond

Add Comment

Respond

Add Comment