Home > CPI, Good One, Rant, TIPS > Hot Button Issue: Rant Warning

Hot Button Issue: Rant Warning


We all have our hot button issues. It will not surprise you, probably, to learn that mine involves inflation. For the rant which follows, I apologize.

Reasonable people, smart people, learned people, can disagree on how precisely the Consumer Price Index captures the inflation in consumer prices. And indeed, over the one hundred years that the CPI has been published such disagreements have been played out among academics, politicians, labor leaders, and others. The debates have raged and many changes – some large, some small; some politically-driven, most not – have occurred in how prices have been collected and the index calculated. If you are interested, really interested, in the century-long history of the CPI, you can read a couple of histories here and here.

If someone is not interested in how CPI is calculated, in how and why changes were made in the methodological approach to calculating price change, then that’s fine. But if a person can’t spend the time to learn the very basics of this hundred-year debate, during which changes were made in the CPI with much public input, not in a smoky back room somewhere, then I wonder why such a person would spend time spewing conspiracy theories on the internet about how the CPI doesn’t include food and energy (um…it does), about how the CPI underestimates prices because it doesn’t account for changes in quality and quantity (um…it does), or about how sneaky methodological changes have caused the CPI to be understated by 7% per year for thirty years.

Recently, the CFA Institute’s monthly magazine for CFA Charterholders was duped into accepting an article that brings together some of the dumbest theories into one place. At some level, the article asks the “interesting” question about whether a consumer price index should include asset prices. Interesting, perhaps, but asked-and-answered: assets are not consumer goods but stores of value. If you are not consuming something, then why would you ever expect it to be included in a consumer price index? You might argue that we should include asset prices into some other sort of index that measures price increases. But we already do. They are called asset price indices, and you know them by names like the S&P 500, the NCREIF, and so on.

Worse, the magazine gives a great big stage to the person who has singlehandedly done more to confuse and anger people, to poison the well of knowledge about inflation, and to stir up the conspiracy theorists about inflation, than anyone else in the world – and all because he is selling an ‘analysis’ product to those people. I won’t mention his name here because I don’t want to advertise his product, but he claims that the CPI is understated by “about 7 percentage points each year.”

That this is being published in a magazine of the CFA Institute is almost enough for me to renounce my membership. It is offensively idiotic to claim that the CPI may be understated by 7% per year, and simple math (which CFA Charterholders were once required to be able to perform) can prove that. If inflation has risen at a pace of around 2.5% per year over the last 30 years, it implies the price level has risen about 110% (1.025^30-1). This seems more or less right. But if inflation had really been 9.5% per year, as claimed, then the cost of the average consumption basket would have risen about 1422% (1.095^30-1).

Can that be right? Well, Real Median Household Income, using the CPI to deflate nominal household income, has risen about 13% over the last 30 years. http://en.wikipedia.org/wiki/File:Median_US_household_income.png But if we use the 9.5%-per-year CPI number, then real median household income has actually fallen 84%. If this was true, we would be living in absolute Third-World squalor compared to how things were in the salad days of 1984. You don’t have to be an economist to know the difference between a slightly-better standard of living and one in which you can afford 1/6th of what you could previously afford. You just need a brain.

Any person who does even rudimentary research on the CPI – say, visiting http://www.inflationinfo.com and reading some of the hundreds of papers gathered there, or perusing the BLS website, or speaking with an actual inflation expert – cannot possibly think that this guy is anything other than a nut or a shill. It is a tragedy that the CFA Institute would publish such trash, and it tarnishes the CFA Institute brand. Let’s hope they publish an apologetic retraction in the next issue.

I also like to point out, when I am in rant mode over this (and, as an aside, let me thank the tolerant reader for allowing me to rant – this allows me to forever point people to this link when they bring up this guy), that if the CPI=9.5% number is right then you must also believe a bunch of other ridiculous things:

First: MIT is in on the conspiracy. The Billion Prices Project, which uses very different methodology from the BLS, figures inflation to be about the same as the BLS does. (Digressing for a bit, I think it’s also interesting that the BPP index has tracked Median CPI much better than headline CPI over the last year, when headline CPI has been dragged lower by one-off changes in medical care prices).

Second: Consumers consistently underestimate inflation, or else are serially optimistic about how it is likely to decline from 9.5% to something much lower. The University of Michigan survey of year-ahead inflation expectations – and every other consumer survey of inflation expectations – is much closer to reported inflation than to the shill’s numbers (see chart below, source Bloomberg). I’ve written elsewhere about why consumers might perceive slightly higher inflation than really occurs, but I cannot come up with a theory that explains why consumers would always say it’s much lower than what they are in fact seeing. Maybe we’re all stupid except for this guy with the website.

michinfl

Third, and related to the prior point: Investors who pour money into inflation-indexed bonds must be complete morons, because they are locking up money for ten years at what is “really” -9% real yields (meaning that they are surrendering 62% of the real purchasing power of their wealth, rather than spending it immediately). We don’t see this behavior in countries where it is known that the official index is manipulated. For example, we know that in Argentina the inflation data really is rigged, and in September of last year long-dated inflation-linked bonds in Argentina were showing real yields of more than 20%. In recent months, the government of Argentina has begun to release figures that are much more realistic and real yields have plunged to around 10% as investors are giving the data more credibility. The upshot is that we have bona fide evidence that investors will base their demanded real yields on the difference between the inflation index they are being paid on and the inflation they think they are actually seeing. The fact that we don’t see TIPS real yields around 6% or 7% is evidence that investors are either really stupid, or they believe the CPI is at least approximately right.

Fourth, and related to that point: if inflation has really being running at 9.5%, then every asset is a losing proposition. There is no way to protect yourself against inflation. You’re not really getting wealthy as you ride stocks higher; you’re only losing more slowly. Since there is no asset class that has returned 10% over a long period of time, we are all doomed. The money is all going away. Especially housing, and real goods like hard commodities – there is nothing you can do that is much worse than holding real stuff, which is only going up in price a couple of percent per year over time while inflation is (apparently) ravaging everything we know and love. There is no winning strategy. Of course, the good news is that it turns out that the U.S. government is being extremely fiscally responsible, with the real deficit falling by 5% or more every year. Right.

I really should not let this bother me. It is good for me, as an investor with a brain, when mindless zombie minions follow this guy and do dumb things in the market. But I can’t help it. The Internet could be a tool for great good, allowing people access to accurate, timely information and the opportunity to learn things that they couldn’t otherwise. It allows this author to come into your mailbox, or onto your screen, to try to educate or illuminate or amuse you. But there is also so much detritus, so much rubbish, so much terribly erroneous information out there that does real harm to those who consume it. And perhaps this is why I get so exercised about this issue: I absolutely believe that people have a right to say and to believe whatever they want, no matter how stupid or dangerous. I am simply aghast, and deeply saddened, that so many people are so credulous that they believe what they read, without critical thought of their own. Everyone has a right to his/her opinion, but they are not all equally valid. There is no FDA for the Internet, so snake-oil salesmen run rampant among their eager marks.

I want my readers to think. If you all agree with me, then I know you’re not all thinking! Look, it is perfectly reasonable to suggest that some minor improvements can be made to CPI. The number has been tweaked and improved for a hundred years, and it will be tweaked and improved some more in the future. It is in my opinion not reasonable to suppose that the number is completely made up and/or drastically incorrect. And that’s my opinion.

Advertisements
Categories: CPI, Good One, Rant, TIPS Tags: ,
  1. onebir
    May 20, 2014 at 4:05 pm

    Is there any evidence that the figures from the unmentionable website are not correctly calculated based on the older methodologies (allowing for discontinuation of some data series)? If not, were Thatcher, Reagan and Volcker idiots for addressing an inflation problem that didn’t exist? (Surely the price of 8 track tapes should have been adjusted down by 75% compared to stereo cassette tapes, and leatherette was a clear improvement over leather 😉

    Isn’t the observation that these figures imply that ‘Consumers consistently underestimate inflation’ consistent with your own belief that monetary policy operates predominantly through money illusion?

    What proportion of improvements to inflation indices over the last 30 years have tended to result in higher published rates?

    Don’t frequent media references to consumer price inflation in low single digits influence consumers answers when they’re asked to estimate changes in the prices of an (ill-defined) basket of goods?

    • May 20, 2014 at 7:11 pm

      Yes, lots of evidence. The main change made at the time that the unmentionable “analyst” claims the big divergence was the change in how housing was calculated – the recognition that the consumption of the house is not related to mortgage rates, or to the value of the house, but closely related to the substitute that is pure consumption: rents. If the ‘old way’ was giving much higher figures, it would imply that (a) mortgage rates were very high (in fact they are very low) and that home prices had advanced many multiples of rents (in fact, until 2006 or so they were strictly in line; they went above that line in the bubble and back below it and are now basically on it again.

      I don’t believe that monetary policy works predominantly through money illusion. I don’t believe money illusion really exists, in the main. What I say is that IF monetary policy could affect real growth, it would be only because of money illusion. But I don’t think monetary policy does much of anything to growth.

      The rental change mainly affected the volatility of the index, not the level of inflation (although it lowered it in the early 1980s and my suspicion is that the ‘analyst’ took a point estimate in one year and assumed it remained constant instead of being ‘paid back’ which, of course, it was).

      I dunno, it seems everything I see in the media refers to the idea that CPI is much higher than people think. But if it’s true that people misperceive inflation as too LOW because of the power of the media, it still doesn’t change the raw mathematics here – there is no question that he is wrong, and mightily wrong.

      • onebir
        May 21, 2014 at 7:41 am

        Thanks for the reply Mike. Apologies for misrepresenting your view on money illusion; faulty memory/ understanding on my part.

        If there are serious errors in the unmentionable recalculated index, I think the best way to debunk it might be to calculate it correctly (or with more reasonable assumptions) & document it carefully. Not a casual endeavour I know… I’d certainly be keen to see this; in particular the effects of hedonic adjustments & the chain-linking (tho I think you identify the latter at around 0.3pp above).

        This however:
        ‘The amount of personal debt would have to be, per year, 6 times the income of the person if it is going to replace the 84% by which incomes now fall short of a constant standard of living. Do you know anyone who borrows several hundred thousand dollars PER YEAR?’
        creates something of a straw man; living standards haven’t been maintained purely by assumption of debt, many other factors have contributed:
        – declining savings rates
        – increases in two earner households
        – increased working hours, reduced holidays, increased secondary employment
        – decreasing fertility
        – increased social transfers (for many)

        A shift from 1.1 workers per household to 1.8 (say, not sure if these are reasonable figures for the US) on its own is enough to offset a 64% decline in real earnings by my calculation…

        Re the media, might there not be a recall bias occurring on your part? Your knowledge of the data may be making the references to higher rates much more memorable compared to the many non-controversial reports based on the official stats every month, and thus over-represented when you try to recall media reports…

      • May 21, 2014 at 10:06 am

        Well, the BLS calculates the index correctly, and essentially all of their calculations are public. Indeed, I can’t imagine how he can calculate any index at all without using the BLS data, unless he’s simply assuming a spread. I’m pretty sure that’s what he’s doing, because otherwise you would surely see dramatic swings in the spread.

        Hedonic adjustments turn out to be a wash, because they tend to increase the inflation registered for housing (because the housing stock declines in quality over time due to aging) and decrease it for some much-smaller categories. The net result is approximately zero. Another case where one of the people who rant about hedonics could save some breath by doing 10 minutes of actual research!

        The change in workers per household would be an interesting theory – good thought! – except that you can see exactly the same trend in real personal income, which is done based on the number of people rather than the number of households. (In any event, the number of workers per household hasn’t changed appreciably since the early 1980s, although it increased a lot in the 1960s and 1970s as women entered the workforce). The basic logic problem remains: in 1984, if a person making $30,000 was just breaking even, then to have the same standard of living that person would today need to be making $456,000. Two people living together can live more cheaply apart, but I’m not sure that adding the second worker at $60k plus the $60k the original worker is making would make much of a dent in a household that now costs what…$750k for two of them? The math is SO wrong that you just can’t explain it away. He would have done better to make up a smaller spread, like 2%, but obviously he can’t do math (or, more likely, doesn’t care as long as conspiracy theorists keep paying him for the ‘research’).

        Re’ the media, you may be right. I haven’t done any study so it may be that I’m subject to confirmation bias. Or maybe you are. We can’t tell, without a thorough research project – and I assume neither of us have time for that! But in any event it only matters to settle the question whether inflation perceptions are biased higher or lower, and THAT depends on whether the real number is 7.5% or 2.5%. Seems we ought to settle that question first, eh? (Of course, I believe it IS settled!)

        Thanks for the exchange!

  2. May 20, 2014 at 5:07 pm

    Although I do not believe that the CPI has been understated by a compound 7.5% per year, I DO BELIEVE that it seriously underestimates the true changes in the cost of living for the average American. I will enumerate my reasons and examples below, but first let me observe that the reasons the average family isn’t living in “third-world squalor” is that they have 1) taken on far more personal debt (credit cards et al) than in prior times and 2) they are buying less-expensive alternatives to the things they used to buy, creating the effect of price stability. But substituting hamburger for steak or rice and beans for hamburger does not mean that the prices of those meats have not risen. It just means that people choose to buy less expensive foods out of necessity.

    As for a “true” inflations index, this is an impossibility because inflation is very much dependent on the individual (or family). E.g., a young commuter in Los Angeles may use two to ten times the amount of gasoline that a retiree uses; whereas a retiree may spend 50 times more on medical costs.

    But there are items that virtually everyone purchases and they are energy (gasoline, electricity, etc.); food; clothing; medical; and shelter. Here are some recent prices increases for typical consumer purchases:

    Gasoline – up 145% in past 5 years or over 19% per year compounded
    Eggs – up 101% in 2 years or 42% per year
    Dryers ice cream – up 33% (same price; pkg reduced from 64 to 48 oz)
    Tuna fish – up 29% (same price; pkg reduced from 7.125 to 5.5 oz)
    Water – up 105% (rate by cubic foot in Grover Beach, California)
    Nursing license – State of California – up 75% (this may seem like a red herring but my wife is a nurse and I use this as a proxy for substantial increases by state and local governments in a variety of “fees” which don’t require voter approval as do tax hikes)
    Lee’s blue jeans – up 19%
    Our personal prescription medicines – up 12%

    I don’t have a billion items like MIT but I think that the above is a start at debunking the 2% myth.

    Have read you for years and enjoy your insights … but please don’t become an apologist for the Fed or the BLS – they both have axes to grind which are not in the interests of the average citizen.

    • May 20, 2014 at 7:19 pm

      The amount of personal debt would have to be, per year, 6 times the income of the person if it is going to replace the 84% by which incomes now fall short of a constant standard of living. Do you know anyone who borrows several hundred thousand dollars PER YEAR?

      The BLS does not assume that people substitute hamburger for steak. This is a myth. What they do is very much more subtle, and amounts to an effect of about 0.3% per year. Max.

      Everyone’s inflation is different, that’s true. But no one ever claimed that CPI was what EVERY person experiences…just the average. Same with Unemployment. You’re not 6.5% unemployed, but on average we are.

      Taxes are not consumption.

      You selected a number of items that have seen increases in a short period of time. Can you find anything – let’s say 5 things – that have increased in value by 1400% in the last thirty years? So they are now 15x as expensive as in 1984. Try. Okay, so you think inflation is 5%? Find a bunch of things that are 4.5x as expensive now as in 1984. What you are doing is a classic cognitive error (no offense), and covered in my article – you notice what goes up. But look at what those eggs have done in price since 1984. Look what gasoline has done in the last 30 years. Look at blue jeans. Etc. Apparel in particular has gone nowhere but down in price since the early 1990s. Blue jeans have risen in price recently – I’ve documented the recent acceleration in apparel that the BLS has indicated by the way – but for the prior 20 years they did nothing. Go look!

      I am not an apologist for anyone. But I AM an inflation expert and I’ve spent years poring over the hundreds of items, methodologies, research, etc. I know this number better than any Wall Street economist and I can tell you – this is about as good a number as anything the government is ever likely to produce! And the BPP tends to support that notion as well.

  3. Lee
    May 20, 2014 at 5:08 pm

    Thanks for the rant/refresher.

  4. Jeremy Fletcher
    May 21, 2014 at 9:10 am

    Have you written a letter to the CFA Institute, or would it be unprintable?

    • May 21, 2014 at 9:55 am

      No. I commented on the site where the article is posted online. Hopefully someone will see it. I need to just let go. 🙂

  5. Frank R.
    May 21, 2014 at 10:42 am

    As I read this, the following came to my mind (you are not old enough to remember the radio series): “Who knows what evil lurks in the hearts of men? The Shadow knows.” Hmmm? I wonder why I made that association? Good rant. Thanks.

    • May 21, 2014 at 12:02 pm

      Ha! Good one! Gee, no idea how that’s relevant, though. 🙂

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: