If that title has you confused, than you are probably not a fan of the CPI ex- food and energy, occasionally referred to as the core inflation rate. That’s the measure some Economists have been using to track the rate of inflation. It’s a foolish game played by those whose grasp on economic reality is tenuous at best.
Ostensibly, removing the more volatile elements of inflation data points avoids having a single outlying month disrupt data. Some of the more numerically literate of you might note that a simple moving average would do the exact same thing, yet allow any simultaneously rising prices to be revealed for what they are.
For whatever reason, some choose to ignore this approach.
Instead, they select the “ex-” methodology of looking at inflation “ex-”inflation. This “ex-” method ignores too many inconvenient facts, i.e., that the CRB Index has been in a strong uptrend since October 2001. Yet despite 4 years of rising prices, the core rate has been remarkably stable. One wonders what the appeal is of such a misleading indicator.
Mind you, this is not the first time the Dismal set has purposefully shifted inflation data downwards. As The Economist reminds us “when oil prices surged in 1973-74, then Fed chairman, Arthur Burns asked the Fed’s economists to strip out energy from the consumer-price index (CPI).” This was to get a “less distorted measure of inflation.”
Unfortunately, they couldn’t stop with just oil – food prices were stripped out too, followed by used cars, children’s toys, jewelry, housing and so on, until around half the CPI basket was excluded because it was supposedly ‘distorted’ by exogenous forces.”
It is no surprise that those who have been overly reliant on the core rate have been unpleasantly shocked recently. The “ex-” group insisted the Fed would pause; after all, why raise rates, if there is no inflation (not once you back out all the inflationary data). Their distress at the most recent hike is directly proportional to their failure to understand the difference between smoothing a data series to reduce volatility, and simply removing inconvenient data that suggests something one does not like. If that reminds you of the recent shenanigans of the Conference Board with their LEIs, than good – you have been paying attention.
Those who live in a seasonally adjusted, hedonically altered, optimized world have to occasionally confront the unpleasant reality of a universe that doesn’t care for their artificial constructs. Ignoring energy – the inflationary data in the CPI – is less than pointless; It shifts the focus away from exactly where it should be: On the part of CPI that has been rapidly increasing in price.
Previously:
The History of Inflation ex-inflation (September 23, 2005)
The Core CPI is useful when juxtaposed to the total CPI, to measure relative rates of change. The disagreement is regarding interpretation.
Ratio of ROC Core to ROC Total. Is it positive or negative, and how far from 1 is it.
What does this value mean for monetary policy.
That’s the way to look at it, no?
Great post.
Any core measure of CPI is supposed to be useful for monetary policy purposes, not because it is an interesting statistical fact!
Central bank can only influence (via the interest rate) that part of inflation that is demand-driven – ie. “core”. Most central banks, therefore, very appropriately focus on “core measures”. Higher food prices are NOT demand driven. An oil shock due to a hurricane is NOT a demand-driven phenomenon. Some central banks even exclude (very appropriately, in my view), the effect of indirect tax changes, regulated prices, subsidies and so on.
Do we end up with a measure that excludes too much? Perhaps. But for monetary policy purposes it makes sense to try and capture what you as a central bank are in a position to influence (demand pressures on prices) rather than the whole CPI. It might be imperfect, but it’s far from a stupid thing to do.
CPI is has been understated for over a generation for good reason. Entitlements– the largest part of the Fed Budget– are adjusted each year for inflation. Imagine what the deficit would be in, say, 2010, if we used a truly accurate measure to adjust for increases in the cost of living.
“oh come on, these are just bureaucrats with green eye shades trying to tweak the formula to make it more accurate!”
OK… but how frequently have changes been made that increased CPI?
This is no conspiracy theory; it’s simply a fact that each and every budget reflects competition for limited resources. entitlements are already such a huge number that ANY tiny adjustment downward makes room for other programs. and the cumulative effect over a generation or two is profound.
There are some very smart folks out there who have estimates of where they think CPI should be today. but even if they are correct, the issue is academic. there is no way to provide benefits compounding at higher rates.
jw
Pasquale:
I pretty much disagree with you across the board — data should be neutral, and the interpretation of it is where shading or coloring occurs.
CPI is simply bad data . . .
But what if the total CPI data isn’t really neutral. What if it is including things that aren’t really inflation (or at any rate, a price increase that the Fed can do something about). Remember inflation is a rise in over all prices. As I understand it, the reason food and energy are taken out of the core rate isn’t just so the data isn’t disrupted. They are taken out because things that affect the supply and demand of those items tend to be unpredictable, which means that price changes are not likely to be related to anything affecting the overall price level.
If you are saying that rising oil prices might increase the risk of (cost push) inflation then I agree with you. I assume that the Fed would as well (in which case I’m not sure we have that much to worry about). But if you were saying that the rise of oil prices is inflation, then I would disagree.
If the recent rise in oil (and other energy products) were considered inflation, then to be seen as effective inflation fighter, the Fed would have to raise rates enough to push the cost of energy down (at least enough so that the entire CPI would stable). Doing this wouldn’t really make sense though, since the recent rise in energy is almost certainly caused by specific changes in supply and demand. It’s hard to see how tightening the countries money supply will make the Middle East more stable, make China grow slower, or bring our oil refineries back online more quickly. It’s not the Fed’s job to deal with the price of oil, just any inflationary effects that spin off it. So I don’t see why oil (or the other volatile items not included in the core rate) should be considered inflation.
Great post.
The numbers are cooked even more with hedonic adjustments. (I see that your blog has already written on this… )
http://bigpicture.typepad.com/comments/2005/06/something_fishy.html)
Its frustrating. I want to buy TIPS (Treasury Inflation-Protected Securities), but I think why bother given the inflation rate is being set so artificially low. Although they may still be a safe haven.
In addition, I feel sorry for social security recipients who are dependent on the government’s inflation number.
The CPI is also being distorted by the manner in which housing prices are calculated, via imputed rent. Since the housing boom/bubble has kept rents artificially low in many areas, this tends to estimate the rate of housing price increases as low at precisely the time that it is actually high.
Not clear how this should be properly calculated, given that houses are normally paid for over a period of decades, but imputed rent ain’t it.
Also, does anyone here know how healthcare prices are factored into the CPI?
Higher food prices are NOT demand driven.
That’s odd because when my grocer has to pay higher costs in order to heat his building he usually demands that I pay more to cover it
Not clear how this should be properly calculated, given that houses are normally paid for over a period of decades, but imputed rent ain’t it.
You could take the average house price and the average mortgage rate, multiply it by the number of homeowners and factor it in that way
It’s not the Fed’s job to deal with the price of oil, just any inflationary effects that spin off it. So I don’t see why oil (or the other volatile items not included in the core rate) should be considered inflation.
Wouldn’t it make more sense for the fed to fight prices we can’t control than those we can? If the price of a computer or TV goes up we can delay the purchase. If the price of food does we can’t so it would seem that (with averaging as Barry suggests) the Fed’s focus should be on forced prices and not discretionary ones
I agree the CPI is intentionally manipulated to keep government payouts under control.
If the Fed’s main goal is to prevent the average person’s spending power from going down, then it seems energy prices have become very important in the calculation and will almost certainly become even more important in the future. As we spend more and more for energy, which we cannot really do without (maybe to a certain degree it can be conserved), this most certainly prevents us from spending on other items which are measured. Therefore, higher energy prices will have an opposite effect on the CPI. I am sure this will become more and more evident as we see energy go through the roof, which is almost a certainty given expected future demand and supply.
I amm certain based on personal experience that prices are going up much more than the measured inflation rate. Monkeying around with the numbers isn’t helping those on a limited income one bit. It does help those with money in the stock market, though. So this is really just more of the same and what should really be expected. The rich will get richer and the poor with get poorer. Has it ever been any other way?
Education, determination, and investing are the key to moving from the poorer statistic to the richer statistic. That’s just the way it will be. No one really believes that politicians will ever change this cycle. Do they?
So expect them to continue to lower the CPI through artificial reasoning for changing the way it is measured.