The Problems with Reported Inflation

The CFA’s bi-monthly magazine includes an article on the gradual mis-reporting of inflation that’s taken place over the years (“Double, Double Toil and Trouble”). Criticism of how the government calculates inflation is not new, and I included a chapter on this in my latest book Bonds Are Not Forever. For a start, it’s hard to take seriously an index which assigns a 24% weighting to Owners Equivalent Rent (OER), a wholly unsatisfactory solution to the problem that owned housing represents an investment that happens to provide a service (which is shelter) whereas the CPI seeks to measure the consumption of goods and services (including shelter) but not the return on investments (such as real estate). OER relies on estimates of what owned housing would rent for if rented, a concept with which very few homeowners have any familiarity. It’s just one example of the white-coated statisticians in their inflation laboratory concocting mathematical potions that have no relevance to the outside world.

John Williams runs ShadowStats.com, and he’s probably one of the better known critics of current practice. One has to acknowledge that the inflation critics are something of a fringe group, and at their most extreme mutter about a widespread conspiracy to cover up the Federal government’s impending bankruptcy. I can’t say I align myself with such extreme views, but nonetheless I do believe that beating inflation as reported is a dangerously low hurdle for an investor to set when planning their retirement.

In the 1990s the calculations were altered to allow for substitution and quality improvements. Both of these can make sense and need not be controversial as long as the user understands what the resulting altered figures mean. Substitution recognizes that the basket of goods and services from whose fluctuating price the CPI is derived changes in real life, and therefore allows that the CPI should reflect these changes. Consumers shift from, say, steak to chicken when relative prices change but also bought fewer vinyl LP records when CDs became available. Clearly, not changing the weights to account for this would mean that the cost of horseshoes would still be a component.

Quality improvements are more subjective in my opinion and rarely seem to include quality deterioration. Commercial flying is a perfect example. When I describe some of the subtleties of calculating inflation in presentations and note that the inflation statistics have incorporated a modest quality improvement in flying in recent years (due to easier cancellation terms) the audience typically laughs at the absurdity of the notion. Longer security lines, poorer food and sometimes surly flight attendants may not resonate with the statisticians who measure such things but are assuredly part of the flying experience for most users of inflation figures.

John Williams gets to the heart of the matter when he argues that in combination, substitution and quality improvements have altered the CPI from measuring the cost of a constant standard of living to the cost of a constant level of satisfaction (constant utility is the economic term). Although the difference may seem trivial don’t be fooled; since living standards generally rise over the time you care more about the cost of maintaining a constant standard relative to your peers and the larger economy. If your lifestyle includes buying the latest ipad you want to maintain that standard, and falling ipad prices (which don’t occur in reality but do occur for inflation statisticians once new model quality improvements are factored in) may give you more utility, but utility isn’t always fungible and you may not have asked for it.

It also occurred to me that in the debates among academicians and government departments about how to calculate inflation, the beneficiaries of the lower inflation camp are invariably well represented. Lower inflation flatters the government’s finances through reduced cost of living adjustments on entitlements as well as tax-bracket creep (since the income bands at which higher tax rates kick in are adjusted up more slowly than they would be otherwise). The tangible and political benefits no doubt invigorate those arguing for a lower-resulting methodology while the benefits to those sincerely advocating a higher-resulting methodology would seem to be far more prosaic.

It’s one more contributing factor to increasing income inequality (since those most exposed to the vicissitudes of inflation calculations rely disproportionately on entitlements and welfare) but it will be no surprise that different inflation calculation methodology has never been much of a hot-button political issue. For investors, you take the world as you find it. Fixed return securities such as bonds provide scant compensation for inflation risk. Someone asked me just the other day what I thought was the best protection against a steady increase in inflation (whether it’s inflation as measured by the government or simply the higher version as experienced). The best answer we have is to invest in equity securities whose underlying businesses possess built-in protection in the form of pricing power. Stable companies with consistently high operating margins; midstream Master Limited Partnerships (MLPs) who often own fee-generating assets whose pricing is inflation linked. These are two of the best forms of defense an investor can adopt. While it’s probably a stretch to assume a government conspiracy, it’s hard to identify participants in the debate whose interest is to err on the side of higher (i.e. more conservative) inflation calculations.