Friday, July 23, 2004

Inflation: "Experts" Dodging Questions

I've talked at some length about how the CPI statistics do not reflect true inflation. So what happens when an "economic expert" get's asked a question about the discrepency between inflation as experienced "in the real world" and inflation as measured by "imaginary world" of the CPI? The "expert", a MSNBC economic commentator, does a lot of hand waving and dodges the question. See for yourself.

What low inflation?
I live in Los Angeles. They say that the CPI for the area is around 3 percent. Just go buy groceries, insurance, or eat at a restaurant, actual prices in many enterprises locally have increased 10-20 percent. How can we believe that prices are really going up only 3 percent when the store prices for just about everything has risen over 10 percent?
Phil M. -- Los Angeles

It’s a little like the economist who has one hand on the stove and the other in a bucket of ice. He’ll tell you that, on average, he feels pretty good.

And that’s all the CPI statistics show: an average. The Bureau of Labor Statistics collects monthly data from 87 urban areas throughout the country, including about 23,000 stores and businesses (for prices of goods and services) and some 50,000 landlords or tenants (to calculate rents).

Aside from big regional differences, price change vary widely from one category of products and services to another. Since Jan. 1998 (through May of this year), for example, food prices overall were up 16.5 percent. But medical care services were up 29.9 percent. Apparel prices, on the other hand, fell by 4.5 percent during that period. And within each category, prices changes vary as well: beef and egg prices have risen sharply in the past few years, but coffee prices have actually fallen (not counting Starbucks).

You’re also correct in your assumption that prices in Los Angeles are rising faster than the national average. Since Jan. 1998 (through May 2004), prices in LA have risen 20.4 percent, while the national average is up only 16.6 percent. It turns out our national economy is very regional -– especially when it comes to the cost of living. Just ask anyone from who’s moved recently from the Midwest to the east or west coasts and gone house hunting.

Then there’s the matter of how much weight to give each item in calculating the average CPI, which the BLS calculates based on overall (there’s that average again) spending levels. But few people, if any, spend exactly these “average” amounts in each category. So if you spend more of your money on clothes than I do and don’t have college tuition bills to pay, you’re personal inflation rate will be a lot lower than mine. (Here’s how the BLS weights each item.)

So why bother crunching all these numbers and coming up with an artificial average that has little connection to real life spending by individuals? Because these averages are useful in making judgments about changes in the overall U.S. economy – especially in monitoring changes over time.

But they aren’t meant to track how your personal spending has risen or fallen. The BLS collects millions of bits of data each year, but even their computers can’t keep up with the individual spending habits of nearly 300 million Americans. Nor, it’s safe to say, would we want them to.

This is a completely dishonest handwaving answer. First of all, The Answer Desk's commentator brings in the red herring of regional variations: "It turns out our national economy is very regional...". No the reader's question stated quite specifically that he was asking about the specific region of LA and that he'd already figured out that number - 3%. As it turns out it is the reader and not the disingenuous desk help person who is correct.

If you go to the site and look up the CPI data on Los Angeles, you can access it here, you can reproduce the data the reader is asking about. All you need to do is go to MORE FORMATTING OPTIONS and select 12 Months Percentage Change and also select the "make a graph" option for your own viewing, and then scroll down the page until you see it. It will confirm that the CPI for LA indicates that the yearly non-seasonally adjusted CPI increase for LA was less than 3%.

The Answer Desk tries to confuse the issue by bringing in price increases over a larger period of time: "Since Jan. 1998 (through May 2004), prices in LA have risen 20.4 percent, while the national average is up only 16.6 percent." Again the reader wasn't asking about national pricing discrepencies. They had already correctly identified that the local change was less than 3%, and they weren't concerned that the price increase IRL were higher than some national average - just their specific city CPI rating! And the price increases the reader was asking about were specifically with regards to recent price increases, and not some compounded six or seven year overall price increase.

Then The Answer Desk once again throws in a red herring: "Then there’s the matter of how much weight to give each item in calculating the average CPI, which the BLS calculates based on overall (there’s that average again) spending levels. But few people, if any, spend exactly these “average” amounts in each category. So if you spend more of your money on clothes than I do and don’t have college tuition bills to pay, you’re personal inflation rate will be a lot lower than mine."

There are more important factors that The Answer Desk is dodging. For instance the practice of using rents instead of property price increases to measure housing cost increases. The problem is that the housing price to rental income ratio is at an all time high, in part because of an increase of rental units and home-ownership simultaneously. Then there is also the issue of "quality adjustments". These hedonic number fixes try to tell consumers that because they've paid more for getting "better" stuff they haven't paid as much, so that doesn't count into the price increase indexes even though the money in real terms does leave the wallet. Also The Answer Desk makes some disingenuous claims: "but coffee prices have actually fallen (not counting Starbucks)". Well yes commodity coffee prices have fallen, but even The Answer Desk has to admit that the price of coffee as paid by the average consumer has not fallen. So since it's supposed to be a Consumer Price Index the fact that coffee commodity prices have fallen shouldn't matter at all. In addition, Greenspan keeps on talking about energy price increases as temporary. Well gas hasn't declined that much from its highs. I wonder if in mid-2005 if oil is still about $40 pb will Greenspan and the government still be counting energy price increases as "temporary".

The Answer Desk disingenuously suggests that the problem is that the readers spending patterns don't match that of the computed basket of goods within that the CPI uses. Well that's bad statistics, mathematics, and bad economics. The mean or average of a distribution isn't enough usually to tell you what is going on. A necessary number to use often is that of the Standard Deviation. The standard deviation is the "spread" that the distribution falls within. Ideally a CPI that actually worked would have a distribution and a mean and a standard deviation. So if the mean is say 3% and the standard deviation should happen to be 1% that means that roughly 68% - two thirds - of the distribution is within 1% of 3%. That means over two thirds of distribution lies within 2-4%.

But here the reader is talking about costs that are somewhere between four or five times that of the CPI mean suggested by the BLS data. The CPI figure for yearly non-seasonally adjusted change in LA is less than 3%. Even if you suggest that the standard deviation is 3% - and it could be scandalous to have an error bar as big as the actual value in hard core scientific values - that would mean that over two thirds of the people in Los Angeles fell within the 0-6% inflation range.

If they fell within two standard deviations away, if we considered the standard deviation to be 3%, that would mean that 95% of the people in LA should fall within -3% to +9% inflation. Yet the reader is talking about inflation that is twice that of the upper end number. And that's assuming a standard deviation as big as the actual value itself, which itself is an outrageous proposition. If I walked in and submitted a paper for peer review and it had error bars as big as the mean value in my measurements, they'd laugh and tell me to stick it where the sun don't shine. What's wrong here is not that the reader has some obscure or bizarre spending pattern, but that the CPI number is completely useless as a mathematical or statistical tool.

Yet The Answer Desk completely glosses over this point. Instead he suggests the CPI figure doesn't have much to do with real life, and not to worry about it. Yet monumentous and sweeping policy decisions and billions of dollars are transacted all the time on the basis of such a number. Isn't that break-taking the arrogance of how the "experts" explain way the difference between the real world and their numbers, and then gloss over the screwed up assumptions in their numbers?

The CPI is completely indefensible. This doesn't stop it from being one of the most often used and important of the numbers that guide decision making at all levels throughout the country. To cut The Answer Desk at MSNBC a break, they do acknowledge that for all intents and purposes the CPI number doesn't have much to do with real life, and they have a standard but well written answer for how the Federal Reserve deals with money supply:
OK, all sounds good. My question is, then, what are the exact mechanisms the Federal Reserve and the U.S. Treasury use to expand or contract the money supply? It has to be more complicated than, “Well, they simply print more money and put it into circulation,” because, who would get this new money then? It can’t just be given away! I have heard and read that money supply growth occurs in two ways: (1) Banks are allowed to lower the percentage of deposits they must retain – this allows them to lend more money as a function of existing deposits; and (2) the U.S. Treasury (or is it the Federal Reserve, or some arrangement of the two?) actually buys back outstanding, previously purchased, U.S. Treasury bonds and bills, which puts more money into the system than existed previously.
Michael G. -- Hillsboro, Oregon

You’ve got the mechanics of the Fed’s job down cold and explained very simply. (Hey, would you mind filling in for us when we go on vacation?)

Of the two levers you describe, by far the Fed’s favorite is open market operations -– where it adds or drains money by buying or selling Treasury debt that's already been issued and is trading in the "open market." Every six weeks or so, the Fed's Open Market Committee meets to give the Fed's New York trading desk its marching orders. When it buys Treasuries, for example, the billions in cash it uses to pay for that paper is then available for banks to lend. So the money supply in the banking system (and the U.S. economy once that money is lent out) goes up. Reverse the process –- sell securities -– and money gets sucked out of the banking system. Seems like a foolproof way to manage the economy.

The problem is that the system was designed and set up early in the last century –- when most lending and borrowing was channeled through banks. Alas, that’s no longer the case. Take the mortgage market, for example, which is dominated by two federally-sponsored entities, Fannie Mae and Freddie Mac. These two institutions lend money to home buyers, bundle those loans as mortgage-backed securities, sell those securities to investors and then lend the proceeds to the next home buyers. This credit creation process takes place outside of the Fed’s control.

There are other vast pools of money that are outside of the Fed’s reach. Money market mutual funds take in billions of dollars worth of deposits and buy various forms of paper -– in effect, lending that money out -- at which point it goes back into the system. The derivatives markets also “create” money. An option (the right to buy or sell a stock at a fixed price within a set period of time) is simply a piece of paper created by an investor willing to place a bet on the direction of that stock’s price. As soon as that piece of paper starts trading, it has a monetary value over which the Fed has no control. And consider the vast wealth that has been created in the housing market. As homeowners tap that wealth with home equity loans, they're monetizing their paper profit. The creation of all of that “new money” is also outside the Fed’s control.

Strictly speaking, this expansion of credit and paper wealth is not the same as printing money. But it has essentially the same impact on the economy -– it increases purchasing power. And, since consumer spending represents roughly two-thirds of U.S. economic activity, increased purchasing power has a much great impact than the relatively few dollars in the banking system over which the Fed exerts direct control.

Even if the Fed were granted vast new powers over any and all dollar-denominated transactions (now there’s a scary thought), it still wouldn’t have complete control over inflationary pressures. For that, we’d have to have a global central bank; much of the current inflationary pressure is coming from outside the U.S. Oil prices, for example, are rising because global demand is approaching production capacity -- or at least oil traders are convinced that’s what’s happening. China’s booming economy is sucking up excess capacity for raw materials like steel and copper, which is pushing up prices of those commodities. No matter what the Fed does, these higher prices cut into the purchasing power of every U.S. dollar.

It’s not that the Fed is powerless – far from it. But the Fed’s control over the money supply – and the U.S. economy – is often overstated.

The reason why I'm quoting this is that it shows that The Answer Desk is in fact capable of factual knowledge and analysis skills. They've answered the question informatively, clearly, and entertainingly. So why didn't they answer the CPI-inflation question the same way without dodging it? I would suggest that deep down the author of this piece understands that the fundamental assumptions of the CPI are flawed. However he cannot admit this publicly without putting himself at odds with the establishment. So the differences between reality and observed facts are explained away with various hand-waving and incorrect mechanisms.

Now expand this by ten thousand times. It is not just one journalist or economist that is failing to take on the flawed assumptions and figures within the CPI, it is the whole government. Either they find it convenient to pretend the CPI reflects real inflation or they find it expedient to act as such, because for instance real inflationary adjustments would massively increase the cost of social entitlement spending. Social Security is indexed to the CPI and even a small upward adjustment might "break the bank" by forcing out much higher benefit payouts.

But even though I can imagine the reasons for going along with such a charade, I cannot condone it. I don't expect this one journalist to take on the BLS and Greenspan all by himself, but the dearth of people speaking out about this in the economic community especially in academia is shameful. Certainly some have tried, and perhaps they were "shut down" by peer pressure or mechanisms. However it's got to end.

This CPI issue is a very big deal for me because it's all part of the bigger puzzle. The Federal Reserve has been since the Reagan era extraordinarily accomodating to debt and money supply in the long run. This has allowed extraordinary borrowing. That in turn lead to political ideologues taking center stage, because pragmatists dominate only when the need for making hard decisions predominates. All of this has been financed by the current accounts deficit purchases by foreign banks. That has depressed foreign currencies to ours and attracted capital fleeing from our shores to theirs, and both ends of the process have stripped the asset base of America. This has been dealt with politically by extending protectionism to politically crucial industry lobbies further distorting the markets. The under-neutral interest rates have in the long term depressed American economic growth, and this has been masked by fixed GDP figures. It has also caused extraordinary inflation, which has been masked by the fixed CPI figures. All of this in turn has caused extraordinary corruption in corporate management because making money has become no longer about succeeding in a free market, but becoming a niche producer protected by regulation. That in turn has lowered the capital investment return for American assets even more, sending capital fleeing overseas. During all of this the American citizen has increasingly given up personal time and gone deeper into debt in order to maintain their lifestyle, even as the financial basis of our economy has been slowly eroding underneath their feet.

The complete story has not yet been told, but it is absolutely true that without these accomodating statistical abberations that such a process could not go on insidiously underneath the nose of every American day in and day out. If people understood what they were really giving up and what the eventual cost would be, I am convinced they would not accept this state of affairs. Even now some courageous individual in the right place at the right time could salvage the situation. However it seems unlikely to me that Greenspan is likely to have a sudden attack of conscience and act from the courage of his convictions. Greenspan is a smart guy. At some level he has to know all of this. It's just a question of heart, and not one of intellect.


Post a Comment

<< Home