A Different View on Inflation

Here's a reader's counter-point to this weekend's piece on inflation.

Marotta’s argument for inflation being chronically underreported by the CPI is founded on the time-honored principle of selectively presenting data in order to reach a predetermined conclusion.  In this case, the predetermined conclusion is to support the almost taken-for-granted conclusion in past articles that inflation has been either double digits or close to it for some period of time, while the CPI and other inflation indicators, as well as all serious economists, disagree with that assessment.

In past articles, one of Marotta’s tactics used to convince readers of his theory was to conflate the concepts of inflation and currency exchange rates – treating them as one, when in fact they are not.  Marotta has used the US Dollar’s exchange rate movements against the Euro as a proxy for inflation.  It seems this approach is no longer as valid to him as it once was, because the dollar has been essentially flat against the Euro for the last 4 months.  This is the first instance of “cherrypicking” the data – when the USD/EUR rate supports the conclusion, pretend that it is a reliable indicator of inflation.  When the USD/EUR rate does not support the conclusion, ignore it.

The second example of cherrypicking data in Marotta’s article is this gem:

“Take 2007 as an example. Bread price rose 7.4%, gasoline 8.2%, health insurance 10.1%, whole milk 13.1%, eggs 29.2%, but according to the CPI, somehow inflation was only calculated as 4.1%.”

Marotta chooses to cite five individual prices, all of which had higher than average inflation, in order to imply that the CPI’s assessment of 4.1% inflation was far lower than the inflation of real individual goods.  However, this is clearly not the case.  To make this clear, let’s look at the food subcategory.  Marotta cites bread prices increasing 7.4%,  whole milk at 13.1%, and eggs at 29.2%.  However, he fails to mention that the “food and beverages” category experienced only 3.9% inflation overall, actually LOWER than the overall CPI!  If Marotta was not trying to deceive you by only presenting data that fits his predetermined conclusion, he might have mentioned that inflation in beef was just 4.4%, pork was 2.0%, poultry was 5.1%, and fish was 4.6%.  Fresh vegetables experienced 3.2% inflation, and fresh fruit experienced 4.5% inflation.  Processed fruits and veggies had just 3.6% inflation.  “Other food at home”, the category that includes everything from peanut butter and soup to sugar, butter, seasonings, and spices, experienced 2.2% inflation.  If the reader had all the information, he might conclude that Marotta had intentionally misrepresented inflation by selectively noting three of the highest-inflation food items, while ignoring meats, fruits, vegetables, and more.  Incidentally, restaurant food inflation was 3.6%.

What about the big picture for inflation?  Am I selectively citing the food category because that’s the only place I can tell the story I want to tell.  Certainly not.  The cost of housing only increased 3.1% during 2007, less than the overall rate of inflation.  Most people spend a hefty portion of their paycheck in this category.  Clothing actually got cheaper.  Drugs and other medical care commodities experienced only 1.4% inflation.  Recreation?  0.5% inflation – that’s for everything from DVDs and iPods to concerts and baseball tickets.  Communications costs?  From telephones to the internet, -0.9% inflation (or should I say deflation?).

Marotta is also selective in his treatment of hedonic adjustments.  It is certainly true that you can’t buy a 1970s car now because of government regulations.  But even ignoring the fact that this is a problem with government regulations rather than a problem with the calculation of an inflation index, and aside from the fact that even without government regulations there really is extra value in things like seatbelts, this example is not representative of the whole.  What about household appliances like washers and dryers that are more energy efficient than ever?  Or computers that constantly improve in capability without increasing in price?  Well, Marotta does mention computers, but if he REALLY thinks that you need to buy a new computer every single year or it will be impossible to run any software, maybe that explains why life seems so expensive to him!  The fact is that in the vast majority of items where hedonic adjustments are used, the old alternatives are either still available (and cheaper than they were several years ago), or have fallen completely out of favor due to the superiority of a new product.  Today’s iPod might be more expensive than a walkman was in 1990, but that’s not really inflation.  Your flatscreen TV is more expensive than a top-of-the-line tube TV from the 70s, but it’s obviously nicer too, or else you could have gone to target and gotten a TV more similar to the 1970s standard.  Is your cable package more expensive than it was 30 years ago?  Maybe that’s because you’re getting about 200 more channels.

Where Marotta chose not to distort the data, you start to see the holes in his argument. 

“With food, the government adjustments are a little more imaginative. They assume if the price of beef goes up, you will eat less beef and more chicken.”

Is this supposed to sound odd?  Because it sounds like a fact of life to me!  Also if I buy my groceries from Kroger, and then Kroger prices go up faster than Costco, I might switch to Costco.

“According to their logic, what we called a car in 1970 doesn't even qualify to be called a car today. It wasn't fuel efficient. It had no airbags, no power windows, no power door locks, no heated seats, no tilted steering wheel and no CD player.”

I think things like fuel efficiency, CD players, power steering, anti-lock brakes, seatbelts, airbags, and yes even power windows and door locks are worth something!  Does Mr. Marotta disagree? 

In conclusion, if you have just one takeaway from this post, look back up at the numbers for inflation in 2007.  Or go to http://www.bls.gov/cpi/cpid07av.pdf yourself to see the inflation of anything you want.  Curious what inflation was for airline fares or prescription eyeglasses? 1.7% and 2.0% respectively.  Look at any piece of the data that interests you and see what really makes up inflation.  It might feel tougher at the pump or in the dairy aisle of the supermarket, but that’s not the only place you spend money.  The CPI represents AVERAGE inflation over a wide variety of goods and services that people spend money on, and there is a great deal of transparency on the BLS website for those who want to see how the components are weighted and how the calculations take place.  The good news for the American consumer is that in 2007, most things didn’t go up in price as fast as gasoline, eggs, and health insurance!

Comments

[Kyle]

Couldn't have said it better myself.

[Matt]

That Marotta post on inflation was alarmist and wrong-headed. Thanks for responding to it in such a systematic, crushing way.

FMF -- Can that please be the last we hear from Marotta? Please?

[FMF]

Matt --

We'll see. For now, his posts are only on Saturdays, so take that day off if you don't like him. ;-)

BTW, many of the commenters AGREED with him.

[Matt N]

Thank you for posting a response to that ridiculous article. Whether or not you agree with Marotta's premise, the manner which he chose to present and support it was inane at best and downright deceitful at worst.

[Matt H]

Matt N: That's about how I feel about Marotta too. I only agree with his premises about 1/3 of the time, but even then, his arguments are so sleazy that it makes the premise look a lot worse than it is.

To the reader who wrote the counterpoint: Outstanding!

I admit I don't ever read Marotta's posts because they are so long. I did read this one because of the couter-article and also found that he used the facts selectively to make his invalid point. Great response!

[Rod Ferguson]

I'm not sure I'd call that cherry-picking. When you are introducing a concept to an audience that they may not understand, you naturally have to keep the math to a minimum or you'll lose your readers. I can see how you came to that conclusion, however, Marotta did say "Not everything is more expensive. Clothes cost less, thanks to continued globalization. And communications costs less too, along with many other electronic gadgets" so he acknowledged the lower prices in some goods. And his "beef to chicken" example is exactly like my example of "steak to hamburger"; the author may not realize that under the new model, the fact you switched to chicken implies that the price of steak is lower than what it actually is. There is a method to this madness; tracking spending habits is a useful tool to help predict shortages in some areas of goods that are traditionally static (if people are going to switch to chicken from steak, farmers and ranchers may wish to grow less steak and grow more chicken). But we should stop calling it an inflation calculator.

We're getting to that point where one has to ask "So.. what definition of inflation are you using?" to be able to talk about it. Austrians say it's monetary. Keynesinans say it's prices and wages. The BLS says it's purchasing habits. It would only be an issue for economists' intellectual enjoyment if the CPI wasn't tied into COLA increases in many things. Since it is, I believe that the people need to know how thier government is calculating inflation so they may do something about it.

BTW - using the old model, CPI is nearly 12%. Using the new model, it's just over 4%.

[Rick]

My vote is to keep Marotta's posts. We may not agree with them all, but they do provide food for thought. In this case, I actually agree with him more than I agree with this counter-post, though I don't completely agree with either.

I found this posts counter arguments against "hedonic adjustments" wrong on several accounts. For instance, it is true that computers get more powerful every year and that this is beneficial for the quality of life. But the CPI counts this as negative inflation, since to buy 2007's top-of-the-line computer will cost less in 2008 than it did in 2007. But how many people really buy used computers? And you may not but a new computer every year, but you still do every few years.

iPods vs. Walkmans? Sure, it may not be precisely inflation that causes iPods to be so much more expensive than Walkmans, but still, the iPod is the de facto standard for portable music, just like the Walkman was the de facto standard 15 years ago. This may be more indicative of lifestyle-inflation, compared with times past.

I think that's what the key here is. There is a lot of art to measuring inflation. It's not all science. You can't say one viewpoint is more correct than another, because there is a variety of opinions on what constitutes "correct" inflation. How much should lifestyle inflation factor in to the official inflation component? We live undeniably better lives than people did even fifteen years ago (cell phones, iPods, etc). And yet this costs more. What's the answer?

[jake]

"author may not realize that under the new model, the fact you switched to chicken implies that the price of steak is lower than what it actually is."

Huh??? That is factually inaccurate. The price of steak is what it is, they just give it a lower weighting in the basket of goods and purchases.

"For instance, it is true that computers get more powerful every year and that this is beneficial for the quality of life. But the CPI counts this as negative inflation."

Well yes, getting more for your money is negative inflation, just like getting less for your money is inflation. Problem being?

"There is a lot of art to measuring inflation. It's not all science. You can't say one viewpoint is more correct than another."

Painting is art too, but that doesn't mean that one can't say Da Vinci's art is better than your 9 year old niece's.

[rwh]

I don't see how you can calculate inflation without taking hedonic adjustments into account. New products are continually introduced to the market, old products are continuously improved, or old products go away.

You can't compare an ipod to a Sony Walkman (which you can still buy, for as little as $15) because they are not the same thing. You can say they are the defacto portable listening device, but that ignores the fact you can buy similar products by other manufacturers for less than $100 and it continues to ignore the fact that MP3 players are a completely different product from the portable radio or cassette player of 1980.

I like the term "lifestyle inflation" because that describes more accurately what's going on in many cases. Houses are bigger, cars are fancier, commutes are longer and many products such as personal computers didn't exist 3 decades ago. And we don't have to drink a $4 latte. They still sell Folgers in a can at the grocery store and serve it at many diners and cafes.

[Rod Ferguson]

"author may not realize that under the new model, the fact you switched to chicken implies that the price of steak is lower than what it actually is."

Huh??? That is factually inaccurate. The price of steak is what it is, they just give it a lower weighting in the basket of goods and purchases.

--------------

Factually inaccurate? When the model is used to show, in part, that the price of steak is rising slower than it actually is.. then the model implies the price of steak is lower than what it actually is. It's called "substitution bias" (i.e. - the price of steak has risen 20% and the price of hamburgers has declined 10%. Ergo, consumers must have "substituted" away from steak for hamburgers so prices are falling.) Yes, you can say the price of steak is what it is. To "measuring inflation" is to see whether the price of steak is different between the past and the present.

------------------

Well yes, getting more for your money is negative inflation, just like getting less for your money is inflation. Problem being?

------------------

No, that's purchasing power. Using more or less money to buy they same things is inflation. You might want to do a little reading on the CPI past and present and inflation in general, Jake.

[Rod Ferguson]

rwh - And we don't have to drink a $4 latte. They still sell Folgers in a can at the grocery store and serve it at many diners and cafes.

----

I prefer Eight O'Clock Coffee, myself. Only a bit of difference in price and a whole lot tastier. :)

[Jim]

Maybe I overlooked it, but I don't see an author cited for this post. Who wrote this one?

I like seeing the counter points to the Marotta posts. Good job.

Jim

[Jake]

"When the model is used to show, in part, that the price of steak is rising slower than it actually is.. "

except that the model is not used to show that. It's used to show that the price of meat in people's diets is rising more slowly than it would be if people were completely insensitive to price changes. The change in the price of steak is not effected by the change in the weighting of steak in the index.

Re: purchasing power discussion -- you ignore the fact that if $1000 today buys a computer 2x as powerful as $1000 would have bought you last year, a computer similar to the one you would have bought for $1000 last year costs less than $1000. That's negative inflation AND higher purchasing power, not just the latter.

except that the model is not used to show that. It's used to show that the price of meat in people's diets is rising more slowly than it would be if people were completely insensitive to price changes. The change in the price of steak is not effected by the change in the weighting of steak in the index.

-----

Exactly. The model talks to meat - not to steak. The *implication* (as in: doesn't state directly, but wants a connection to be drawn) is that steak isn't as expensive as it is. A useful model to track dietary habits, not so useful to track inflation.

-----

Re: purchasing power discussion -- you ignore the fact that if $1000 today buys a computer 2x as powerful as $1000 would have bought you last year, a computer similar to the one you would have bought for $1000 last year costs less than $1000. That's negative inflation AND higher purchasing power, not just the latter.

------

No, that's a hedonic adjustment, which has already been covered. Inflation is either a monetary phenomenon (more or less money in the supply) or a price/wage phenomenon (higher or lower prices and wages) and purchasing power is how many units of that money are needed to buy the same goods. For example: in an inflationary period, a $10 item may now cost $15 (more money needed) and in a purchasing power decline period, a 10-Euro item made overseas is still only 10 Euros, but we need 15 dollars to buy it. Granted, it sort of looks like inflation, but they are very different things (Austrians will argue that they are born from the same root cause).

[Tom]

There is a problem with hedonic and substitution adjustment that I think most commenters are missing. I agree that using hedonic and substitution adjustment is an equally valid method of measuring inflation. However, the true problem with these adjustments is that the new methods make historical comparison difficult. Why is this a problem? Remember when the 1970s had double digit inflation? We rightly remember then as bad economic times. Today, we're told that times are less than great now, with inflation running about four percent, but, hey, at least it is not as bad as in the 70s when inflation was double digit. The problem is, using the 70s method of calculating inflation, inflation is now more than ten percent. So why are we being told that we are not as bad off as then? The problem is not that inflation is being calculated wrong, the problem is that our government is intentionally deceiving us by comparing apples to oranges.

[rwh]

Tom:

I agree historical comparisons are difficult, but it appears to me the current way of measuring inflation is an improvement over the old way, because it takes consumer behavior and the changes in consumer products into account.

So perhaps, instead of taking issue with the view that inflation is understated now compared to the 70s, (and current economic conditions are worse than we're being told) maybe inflation was overstated in the 70s and things were not as bad then as we thought.

[Tom]

RWH:

Your interpretation is possible, but, in my view, unlikely. The '70s inflation measure matched consumer perceptions regarding the health of the economy, while the current measure of inflation doesn't seem to match the public's view of the current state of the economy [currently, the public view is more negative than the inflation number]. I don't think the inflation measure drives consumer perceptions, or people would not be as negative about the economy now. I think more likely both inflation then and now [even if driven by just food and oil] is intolerable. I think the government is telling us its not as bad as we think because it is an election year. However, the public will trust their subjective impressions over a government statistic.

[Jake]

"So why are we being told that we are not as bad off as then?"

Who is telling you this?

http://news.google.com/news?hl=en&q=%22since%20the%20great%20depression%22&um=1&ie=UTF-8&sa=N&tab=wn

http://news.google.com/news?hl=en&um=1&tab=wn&q=stagflation

http://news.google.com/news?hl=en&um=1&tab=wn&q=%22worst+recession%22

http://news.google.com/news?hl=en&um=1&tab=wn&q=%22not+as+bad%22+1970s+inflation

[Jake]

"The *implication* (as in: doesn't state directly, but wants a connection to be drawn) is that steak isn't as expensive as it is. "

That's certainly not an objective fact, you're now trying to read the minds of the people who work for BLS, which seems pretty silly to me. They're stating something accurate and you're criticising them because if somebody INCORRECTLY interpreted their numbers to say that steak is cheaper than it actually is, that person would be wrong. That's a pretty weak argument.

"No, that's a hedonic adjustment, which has already been covered."

I don't think you read the last thing I wrote, so I'll write it again before leaving this discussion to others: you ignore the fact that if $1000 today buys a computer 2x as powerful as $1000 would have bought you last year, a computer similar to the one you would have bought for $1000 last year costs less than $1000. That's negative inflation AND higher purchasing power, not just the latter.

You can just say "no it's a hedonic adjustment", but that's not an argument. Accounting for hedonic adjustments gets you closer to a real inflation number, not farther, because the portion of price increases that are due to getting different/superior products are not actually inflation. For example, if I went to Best Buy last week and bought a 32 inch tube TV, and next week I go to Best Buy and buy a 50 inch plasma TV for 4x the cost, that does not mean that TVs have experienced 400% inflation. The same logic applies when you look at the whole country buying a changing mix of improving products over a period of decades.

[Jake]

"I think the government is telling us its not as bad as we think because it is an election year."

Common plank in the construction of conspiracy theory, but does it stand up to scruitiny? No.
(1) No changes have been made to the calculation methodology in many many years, so no policymakers who might actually care about the results of the 2008 elections have tampered with any calculations or methods.
(2) Nobody who provides the inputs for the calculation is elected to anything by anybody, they're low-ranking civil servants who go find out how much tomatoes etc cost.

[Rod Ferguson]

Jake -

That's certainly not an objective fact, you're now trying to read the minds of the people who work for BLS, which seems pretty silly to me. They're stating something accurate and you're criticising them because if somebody INCORRECTLY interpreted their numbers to say that steak is cheaper than it actually is, that person would be wrong. That's a pretty weak argument.

-------

The CPI was used to track the prices of certain goods and services. A fixed basket of goods and services. Tracked over a period of time. That basket is no longer fixed, and the tracking is no longer like-for-like. The BLS makes no bones about it - they admit as much. You are correct that it's not the responsiblity of the BLS to have people correctly interpret thier data; that's up to the individual. Hence this discussion. I'll retract my statement and change it to "misrepresenting the cost growth in steak by comparing the price of steak to the price of chicken" rather than "implies that the price of steak is lower than what it actually is."

It seems to me that you agree that the data isn't representative of the historical norm for the CPI; you simply do not think the change makes the CPI a bad indicator for inflation. Others may disagree; I'd hazard a guess that these few posts on the CPI and the concurrent discussion would be most readers' first exposure to what the CPI really is and therefore, valuable. There is a little monetary trick to see how accurate the CPI is to real inflation: track the inflation rates of those nations whose currency is pegged to the dollar. If a currency is pegged, then domestic inflation should be in line with that of the US. The inflation rates of these nations range from 7-10%, far above our current CPI and much closer to the old model.

-------

You can just say "no it's a hedonic adjustment", but that's not an argument.

-------

It wasn't an argument - it's an explanation. It's what "you ignore the fact that if $1000 today buys a computer 2x as powerful as $1000 would have bought you last year, a computer similar to the one you would have bought for $1000 last year costs less than $1000." is called. You said "That's negative inflation AND higher purchasing power, not just the latter." and it's not - it's a hedonic adjustment.

Look, the fact we're printing too many dollars might be bad, but if you call that dollar printing "investor speculation" you'd be wrong. It's called "monetary inflation". I'm not ignoring hedonics, or even stating that hedonics are bad - just that you're calling hedonics "negative inflation" and "purchasing power" when they are not.

[Lord]

One thing Barry Ritholtz points out is while hedonic factors that lower inflation may be measured, those that raise it may be missed. These would be along the lines of lower quality, poorer customer service, less support, increased delays and waiting. These can impair values but are often not directly monetary and therefore may escape measurement.

Steve Miller asked the following questions in You try setting the rules on another blog:
1. A company comes out with a new package size for a product. The size is 50% larger and the price goes up by 25%. Should we view the price of the product as lower, unchanged, or higher?
2. The price of homes is going up, but interest rates are moving lower. One person sells an existing home and buys another. Both homes increased by 20% in price during the prior year. The second person remains in the same home with the same mortgage, living next door to a place that went up by 20%. The third person, who also lives next door, refinances his mortgage to pay off the loan much more quickly with the same monthly payment. Two of the three people are paying no more for their housing, although the home price went higher. One is paying of the mortgage faster. How would you measure the increase in home prices, given the decrease in payments?
3. You are told that a group of 24 other countries has adopted some rules about private accounting. (Forget things like FAS 157 and marking to market -- those are out! International accounting does not follow these rules.) Do you think that this is a good argument for U.S. accounting to follow suit? What if the other countries had different rules about free speech? About voting? About religion and education? Briefly put, does the fact that a group of other countries does something mean that the U.S. should also, without regard to the actual values and merits?
4. Let us suppose that the price of a car in year X is $15,000. A few years later the car has the following improvements:
* Anti-lock brakes -- improving safety.
* Side bars to protect in side collisions.
* Air bags to protect in frontal collisions.
* Fuel injection and computerized sensors to improve fuel economy.
* Stronger and safer tires.
* A life expectancy of 150,000 miles instead of 50,000 for the earlier car.
* Better cup holders, including both warming and cooling.
* If the car still sold for $15,000, would you say that the actual price had decreased or remained the same?
5. You are trying to measure recreation costs for a typical Chicago family that likes sports. The Bulls use their #1 draft pick to choose the next Michael Jordan. To help with costs, they raise ticket and concession prices by 30%. You give up your Bulls tickets and instead buy more tickets to the Bears, marked down after failing to get a Quarterback in the draft and releasing their top running back. You spend the same amount of money to watch football instead of basketball. Has there been an increase in the price of recreation?

My answers:
1) One can be tricky since larger sizes are normally cheaper. It would also be worth comparing with larger sizes if possible to see if there wasn't in fact an increase.
2) Unchanged to first order. Higher property taxes would increase it though.
3) No, but it is worthwhile considering and possibly even reporting on comparable basis.
4) Decreased, but this should be evident on resale. If it still depreciates as fast, than it would not have decreased since those improvements have become obsolete faster.
5) Increase unless the fall in Bears tickets offsets the increase in Bulls tickets. Only being able to afford an inferior product does not mean prices have remain unchanged.

[Bad_Brad]

The bottom line on inflation is it depends on what you're buying. The "market basket" used in defining inflation is set by the CPI but reality is, it's different for everyone. Right now, if you buy a lot of gasoline, groceries, and medical services (i.e. if you commute 60 miles each way to work and have a spouse and five kids), then you are getting creamed by inflation. On the other hand, if you buy a lot of long distance phone calls and high-end electronic devices (computer, plasma TV, etc) (i.e. you are a single technophile who walks or bikes to work and lives 4000 miles from your parents who you call everyday), you are probably saying, "inflation? what inflation?"