The internet and inflation

From Austan D. Goolsbee and Peter J. Klenow:

We use Adobe Analytics data on online transactions for millions of products in many different categories from 2014 to 2017 to shed light on how online inflation compares to overall inflation, and to gauge the magnitude of new product bias online. The Adobe data contain transaction prices and quantities purchased. We estimate that online inflation was about 1 percentage point lower than in the CPI for the same categories from 2014–2017. In addition, the rising variety of products sold online, implies roughly 2 percentage points lower inflation than in a matched model/CPI-style index.

I call this “the gains from better matching,” as discussed in The Complacent Class.

Comments

There does appear to be something wrong with the way most Western governments measure inflation. I would guess politicization of everything has got to the measurement of inflation. But that ought to lead to governments pressuring their wonks to *under*state inflation. Which would imply the internet figure is wrong. I am not sure I would agree with that idea.

Perhaps an unrepresentative set of goods is sold on the internet? Although I would think that so much is sold that there is a pretty good spread these days. On the other hand housing is not likely to be sold on the internet all that often. How much inflation is housing price inflation? Medical care is unlikely to be sold on the internet either.

Perhaps the greater efficiency of the internet is not being captured by the government's figures?

Is this a good thing or a bad thing? Much is made of the Fed's inability to reach its 2% inflation target. Perhaps online sales is one of the reasons. Online sales are more efficient in the sense that prices are lower because costs are lower; but costs to one are revenues to another. Maybe it's losses, not gains, from better matching.

This findings, or other findings shouldn't make anyone reconsider their own real income very much - you already observe the prices relevant to you.
However, ramifications are large for the analysis of aggregate real incomes over time. The much-publisized finding of "no real wage gains since the 1970s for the median worker" are very sensitive to even a 1% bias in measured inflation.

I am not sure how much stock I'd put in this particular paper, but it illustrates nicely the pitfalls of "deflating" incomes at points in time that are far apart.

So increased online sales caused the great recession, right? That's absurd. But so is the claim that low interest rates cause low rates of economic growth and high interest rates cause high rates of economic growth. Economics is becoming a parody of itself. Interest rates fall because of weak demand for credit, and rise because of strong demand for credit. The Fed has little to do with it. Indeed, in a recession efforts to lower interest rates signal an extended recession, exacerbating the already weak demand for credit. As for online sales, the more of them, the fewer sales in the mall, the empty mall signaling a weak economy, exacerbating both the weak demand for credit and the preference for savings over consumption. Gibberish, insight, or parody? I could go on but mercifully I will stop.

I truly have no idea what you're on about. As far as i can see it has nothing to do with anything you wrote in the first post or what I wrote in response

It's rayward, he often makes replies of that nature.

As I say below, I agree that felt inflation can be different between individuals, and even a choice, but I disagree that average punters (a) know their felt inflation or (b) manage it.

Probably most people just treat gasoline prices as (felt) inflation, punctuated by the "news" that "inflation is running at X% this year." Heaven help 'em.

Are you saying that people don't respond to rising gas prices by cutting their usage?

I think you missed a couple things. The first is that gas prices may not actually be the largest component in a family's "felt inflation." The second is that "usage" isn't decided that way. The fundamental choice is "SUV or not," basically. And that will be made with a lot of subjective emotion, modulated by probably wishful thinking about forward gasoline prices.

I think you avoided answering the question.

Gawd. If you could be less fast-and-combative maybe you could add something to these conversations.

We have a problem with overloading of terms. "Inflation" means different things to different people, and given they way the word is sold in various mass and specialized media that isn't going to change anytime soon. Inflation may mean:

1. The strict monetary effect.

2. The subjective feel of price changes.

3. The change of various structured "inflation indexes."

and what I'm suggesting:

4. The actual, but uncalculated quantitative effect of prices changes on consumers.

The way we can add value in any conversation is to recognize all these views, and how they link or don't in specific circumstance.

But "people drive less" is absolutely the minimum (non)answer in terms of thought and contribution.

People will drive less when Gas prices go up faster than the general rate of inflation. Yes, there are numerous factors that make gas usage less elastic than the median good, but that still doesn't change the basic supply and demand function.

"but I disagree that average punters (a) know their felt inflation or (b) manage it. ... Heaven help 'em."

I believe that most "punters" are more observant and smarter than you give them credit for.

You are still not operating on the right level, unless you are trying to argue that "driving less" is all people need to do to manage their response to consumer price inflation?

"4. The actual, but uncalculated quantitative effect of prices changes on [specific or individual] consumers."

5. I just realized I left out supply shocks as things that are called "inflation."

The fact that inflation is a monetary phenomenon continues to be ignored, proving that empiricism has yet to penetrate economic thinking. Did the advent of big box retailers that lowered prices on day-to-day consumer purchases result in deflation? Are increases in college tuition inflationary? Is a 2% Fed inflation target a means of increasing the supposedly beneficial velocity of money or evidence that Satan is the real head of the Federal Reserve?

Truth.

One concern with your comment: Satan gets it (in an evil mode) right sometimes, the Fed almost never.

Couldn't get more than an abstract, so I ask, did they include delivery fees? The delivery stack for on line shopping is short, but warehousing costs are complemented by delivery costs, essentially using the delivery companies to cover warehousing costs that otherwise would show up in CPI numbers.

The implicit deflator has been hovering as 1.5%, a number 2% below that says online goods are decreasing in price, which I doubt.

If you buy a product online, you’re buying the bundle of the product and the home delivery. It’s not clear to me that including the whole bundle price is necessarily better for comparison purposes.

"Couldn't get more than an abstract, so I ask, did they include delivery fees? "

I seriously doubt they include the cost in time and fuel to visit a store as part of the costs. And that cost can be significantly higher than Amazon delivery charges.

To clarify, they don't include delivery costs on normal cost of goods either.

Delivery costs are embedded in price certainly, they do in and out price flow. It is definitely not priced in online, there is a separate fee, and the includes priced service options. The online business was built with UPS and post office, all charging separately.

"Delivery costs are embedded in price certainly,"

No, you are missing the point. It costs time and fuel for me to drive to the store to pick up a part. This cost is often higher than the cost of Amazon delivery charges for one off purchases.

If you are going to factor in the cost of delivery from Online distribution center to a residence, then you also would need to factor in the cost for a person to drive to a retail establishment and then drive home. That would be an Apples to Apples comparison.

It costs time and fuel for me to drive to the store

How much an hour are you charging yourself to drive to the store? How much an hour are you charging yourself to watch Seinfeld re-runs? How much an hour does it cost you to sleep 8 hours and 21 minutes each night? Over and above the bill, what's the cost of a meal at Olive Garden? Is your time reflective of the labor theory of value or the demand/supply curve?

"How much an hour are you charging yourself to drive to the store?"

I value my personal time at $10-15 per hour, however, my wife values my time at roughly $2 per hour. So it probably averages out to minimum wage.

"How much an hour are you charging yourself to watch Seinfeld re-runs?"

That costs me an hour of time, not playing Fallout 4. So, I don't watch Seinfeld re-runs.

"How much an hour does it cost you to sleep 8 hours and 21 minutes each night?"

Married with 4 young kids, so that pretty much approaches infinity.

"Is your time reflective of the labor theory of value or the demand/supply curve?"

Yes.

@Chuck Martel

+1

The concept that "rising variety of products sold online, implies roughly 2 percentage points lower inflation" is ludicrous. Normal market supply/demand dynamics do not and cannot cause "inflation".

The term "inflation" apparently has no objective meaning, even among economists

I am pretty sure that China's opening up and reform has lead to a lot of downward pressure on prices. Everything, or most things, have become a lot cheaper. We would have deflation if the currency was still gold.

So do you agree that China can produce deflation?

As I've mentioned, I think this goes in tandem with good deals disappearing from brick-and-mortar stores. Sure there are sales, but certain classes of efficient products or simply not offered other than online. You can buy what is in stock locally, or you can suffer the delay to get those other products.

An HDMI cable used to be a good example. You can get one for maybe $3 at Amazon, but if you went to Best Buy the cheapest one in stock was $9. It would be Deluxe or improved or whatever.

So I guess everyone gets to choose their inflation. As others above say, felt inflation is different and more a choice than monetary inflation.

(Prius drivers experience different felt inflation than Mustang drivers.)

Liked it so much you posted it twice.

Felicitations!

Don't feel bad Goolsbee's paper shows he doesn't know squat about inflation, either.

Double click. I assume it will be cleaned up later. If only your zero-content comments were as well.

As I've mentioned, I think this goes in tandem with good deals disappearing from brick-and-mortar stores. Sure there are sales, but certain classes of efficient products or simply not offered other than online. You can buy what is in stock locally, or you can suffer the delay to get those other products.

An HDMI cable used to be a good example. You can get one for maybe $3 at Amazon, but if you went to Best Buy the cheapest one in stock was $9. It would be Deluxe or improved or whatever.

So I guess everyone gets to choose their inflation. As others above say, felt inflation is different and more a choice than monetary inflation.

(Prius drivers experience different felt inflation than Mustang drivers.)

BTW, my girlfriend has a fancy water filter for her refrigerator. It used to be that she had to call the plumber once a year and pay $300 to get it replaced. Now we can order it from Amazon for $70. In that case it's both cutting out middle-men and access to an obscure part that wouldn't be at any local store.

(Sub-Zero, if you are wondering why a water filter could be $300)

Walmart has one advantage — it's much larger than Amazon by revenue. Walmart's 2017 sales surpassed $500 billion for the first time, making it almost three times bigger than Amazon. That size enables Walmart to put a much bigger price squeeze on suppliers than Amazon in online shopping, Barrett says.

=========

What is WalMart's advantage in brick and mortar?

On line sales technology, they use it internally more efficiently than a dispersed system like Amazon. Home Depot is good at this They do better matching, just like Amazon, their huge warehouses are optimally matched and priced with the same analytical tools used by Amazon.

They have one key advantage, consumers go batch shopping, it is efficient but Amazon, in particula Jeff, ain't bright enough to decode that, yet.

The Billion Prices Project tracks online prices vs BLS inflation. Pretty close.

http://www.thebillionpricesproject.com/usa/

Comments for this post are closed