There are multiple problems when debates over inflation and deflation break out.  The primary problem is that humans tend to be extreme in their beliefs.  There is too often no room for middle ground.  For instance, I am often taken out of context as a deflationist even though I maintained for the entirety of 2010 that we were likely to suffer disinflation and then earlier this year forecast low levels of inflation in 2011.  Nonetheless, because I reject the notion of hyperinflation or even high inflation I am pegged as the extreme opposite – a deflationist.  This extremism results in losing sight of the highest probability outcomes (which is likely to be neither deflation nor hyperinflation).  It’s great to have conviction in a belief, but it must be tempered by reality and probability.

In addition, humans tend to have very short memories resulting in attentional biases. In the case of inflation we tend to focus on what has happened only just recently as opposed to what has happened around us over the course of several months or years.  For most of us, this involves seeing gasoline signs, stock prices, gold prices or other noticeable prices.  These biases combine to lead most of us to constantly fret about near-term price changes in highly visible prices.  Gasoline prices are rising in the last few months so it must mean that we are on the verge of hyperinflation, right?  Not so fast.

The Cleveland Fed recently devised an inflation index that most readers can likely relate to.  It is called the Flexible CPI.  This index prices the more volatile measures of CPI and if you look at the chart below you’ll likely relate to the price changes better.  Yes, inflation really did feel high in 2007 & 2008 despite the low readings in the core CPI.  But the story was not complete just because flexible prices were surging.  When many were fretting about inflation their concerns were not without merit, however, they weren’t entirely accurate when one pulls back and looks at prices as a whole.

This is where the sticky prices index comes in.  While gasoline prices may be gyrating on a daily basis the price of many other goods and services (such as your rent or mortgage cost) remains relatively constant.  The Cleveland Fed elaborates on these indices:

“Another way to analyze the incoming data is to look at where the price increases are coming from. Bryan and Meyer (2010) separate the consumer market basket into “flexible” and “sticky” prices. Flexible-priced items (like gasoline) are free to adjust quickly to changing market conditions, while sticky-priced items (like prices at the laundromat) are subject to some impediment or cost that causes them to change prices infrequently. As their research shows, sticky prices appear to have an embedded inflation expectations component that is useful in forecasting future inflation.

As is evident in the figure below, the flexible price series is definitely more volatile, and does appear to vary with changing economic conditions. The sticky price series has been relatively stable since 1983, usually hovering between 2.0 percent and 3.0 percent. However, over the past two years the sticky CPI has experienced a sizeable disinflation—slowing from a year-over-year growth rate of 2.8 percent in December 2007 to a low of 0.7 percent in September 2010. Since then, the sticky CPI has edged back up slightly and is now trending at a 12-month growth rate of 1.0 percent. The flexible CPI, which fell to a year-over-year growth rate of -10 percent during the depths of the last recession, has popped back up to a 12-month growth rate of 3.4 percent through January.”

“The flexible CPI is intriguing in that, by design, it is likely to show evidence of pricing pressure ahead of the sticky CPI. However, the series is very volatile relative to its sticky-price counterpart and likely dominated by relative price changes. As a result, inflation forecasts based on the flexible CPI perform rather poorly.

While rapid price increases in a few categories seem to have pushed up the headline CPI lately, underlying measures of inflation are relatively low and have only ticked up slightly in the past few months.”

The study referenced above elaborates further:

“We find that forecasts of the headline CPI that are based on the sticky-price data tend to be more accurate than the forecasts based on headline inflation. Further, CPI predictions using sticky-price data perform pretty well relative to CPI forecasts using core CPI data.6,7. We also find that the relative accuracy of the sticky-price Phillips curve increases as the forecast horizon gets longer. For example, when predicting three months ahead, we find that the sticky-price Phillips curve reduces the RMSE of the forecast only about 2 percent relative to the headline CPI. For the 24-month ahead forecast, the improvement in the RMSE was about 14 percent. The flexible-price measure, at least on the surface, does not seem to forecast well, and it performs increasingly worse as the forecast horizon gets longer.”

This doesn’t mean that flexible prices are always wrong.  But we must remember not to be too biased about near-term highly visible price increases.  There’s more to the story than that.  And as we learned in 2008 a short-term spike in flexible prices can occur just months before a nasty deflationary event.  Given that there are many similarities between 2008 and the current environment it would not be surprising if flexible prices continued to surge into the summer months before triggering their own self correcting mechanism in the form of reduced aggregate demand and slower economic growth.  So while surging flexible prices are important to keep an eye on it’s equally important to remember that flexible prices are only one piece of the overall inflation story.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:


  1. On the surface what is suggested here is correct, but there are some technicalities which slightly alter the usefullness of the two measures and affect the potential outcomes. It was Andy Xie I think who talked about looking at the smoothed affect of flexible prices over a longer timescale. While flexible prices have been volatile there is an underlying trend to them which appears to have altered since 2000, which I would atribute to changes in the way markets work and due to Fed actions. This would suggest a combination of sticky prices and smoothed flexible prices would give the best view of CPI, worryingly the background uptick or downtick in flexible prices might be being ignored.
    Whilst I am inclined to agree somewhat that the 2008 oil price scenario will be replayed, I don’t think the effect will be as significant on CPI. Here I think there were two components to the CPI drop after the oil price peak in 2008. The first was the drop in oil prices, but a major part was due to house prices and equivalent rent costs. What I am suggesting is that there are enough differences in the position of the economy now compared to 2008 that a correction in oil prices might not be as strong and may not impact CPI in the same way.
    The article kind of pitches things right by saying “This doesn’t mean that flexible prices are always wrong” and “it’s equally important to remember that flexible prices are only one piece of the overall inflation story”. The last statement being key as there are a lot of factors outside of oil and housing particularly around the state of economies outside the US which will play a part.Its just important to remember that sticky prices and core inflation are not the only story either.

  2. You are a fool.

    Stop writing this pseudo intellectual rubbish and tell us how much cash have you made from the markets in the past 2 years.

    What a waste of time.

      • Aw, let’s give Paul the benefit of doubt, and assume he really is a child having a temper tantrum, while pretending to be a grown man.

        Cullen could be wrong, but he is clearly no fool, and it is equally obvious that he cares about educating his readers, while apparently expecting little in return, except that we behave civilly.

        The explanation was very valuable.

    • it would appear that Paul Skinner has made no money during the greatest bear mkt rally in history, is quite mad about it, and is blaming anyone and everyone – including TPC – for not “telling him to buy” in Feb-09. Of course, Paul Skinner would not have went long in Feb-09 even if TPC had told him to, but that is another story…

  3. To say that Cullen Roche is a fool is ridiculous to say the least.

    If you do not like is biog go some where else. Its a free world. Considering the amount of information and the questions he answer in a day I think he does over all a hell of a good job. He is concentrating a lot of energy demystifying the monetary system of the US. What is wrong with that ?

    Inflation is related to fiat money and he as a opinion. Even if he made a mistake I don’t think he said he was GOD.

    This is a free site and its very informative.

  4. Very interesting! I wasn’t aware that “Sticky / Flexible CPI existed. It explains a lot of things.

    As to the hyperinflation / hyperdeflation arguments I consider it noise to be filtered out. I’m a trader and only interested in probable outcomes.

  5. Mr. Roche,

    I have a question I hope you will answer.

    I have read your pieces on QE and the monetary system. You say the US gov’t cannot have a funding problem because they just spend/create the money they need. You say the bond market is not a fiscal/funding tool but purely a monetary tool for the control of reserves and thus interest rates.

    However, the gov’t does pay interest on Treasurys, right? So, even if engineering lower rates in the bond markets via QE does not impact their ability to fund operations , doesn’t it impact the amount of money the gov’t needs to create (ie lower rates = lower interest payments = less money to create).

    So don’t lower rates help keep a lid on inflation and give the government more room to create/spend more money? While at the same time, lower rates may spur consumer spending which can grow the economy and justify/balance gov’t spending increases and prevent high inflation?

    Does this make sense? If not, please explain if you would. Thanks.

    • Not sure if I completely follow your question. Govt bonds actually add to net financial assets by paying interest. This increases the deficit. The govt can never not be able to pay its interest on bonds. Lower rates are supposed to reduce the burden on debtors. But it also increases the burden for savers….in this environment, however, the govt has clearly chosen to aid the debtors and punish the savers.

      • What I meant was. If the government simply creates all the money they need, don’t lower interest rates moderate inflation because the government has to create less money to pay the interest on gov’t bonds?

        Since the only possible constraint on gov’t spending is the fear of high inflation, by keeping interest rates low the government can spend proportionately more money on operations and stimulus, and less on interest payments.

        In other words, lower rates allow the gov’t to spend more money while not increasing the risk of inflation.


  6. What is the term for deflation for some items such as housing, and inflation for others such as food, clothing and fuel. I don’t see the flexible vs sticky distinction, as it usually takes time for wholesale prices to get to retail, and prices hikes happen quickly, no matter what. The reason we have house price deflation is that no one is buying. The reason we have food inflation is that everyone must buy…

  7. Dear Sir,

    The core problem with your argument is your reliance on “New CPI” data. You seem to forget or ignore that the definition for CPI (a basket of basic consumer products) is controlled by “the government”. CPI also does not include
    “energy”. The formula for CPI was changed under several Presidential administrations and now (purposely – in my opinion) understates inflation for the following core reason:

    ** To reduce pension, SSI cost of living and (Treasury (I-Bond) obligation adjustments, which in turn keeps Government payments down and obscures the true effects of inflation.

    Please refer to the following article for details:

    “Payments to Social Security Recipients Should be Double Current Levels

    Inflation, as reported by the Consumer Price Index (CPI) is understated by roughly 7% per year. This is due to recent redefinitions of the series as well as to flawed methodologies, particularly adjustments to price measures for quality changes.

    In particular, changes made in CPI methodology during the Clinton Administration understated inflation significantly, and, through a cumulative effect with earlier changes that began in the late-Carter and early Reagan Administrations have reduced current social security payments by roughly half from where they would have been otherwise. That means Social Security checks today would be about double had the various changes not been made. In like manner, anyone involved in commerce, who relies on receiving payments adjusted for the CPI, has been similarly damaged. On the other side, if you are making payments based on the CPI (i.e., the federal government), you are making out like a bandit.

    • As I’ve said many times, income growth and CPI are highly correlated for obvious reasons. If your argument holds true then it means the govt is lying to us about low income growth in order to tell us inflation is also low. So, they lie by telling us we’re poor in order to lie about low inflation? No, your theory doesn’t pass the common sense test….

  8. Dear Sir,

    I believe you miss the point on CPI here. As you missed the point with your (very narrow) argument regarding The Fed’s QE program.

    CPI is a “manufactured and manipulated” government index that I would argue is not indicative of the average household’s cost for a basket of common goods. Because it does not include “Energy” (and I also believe it does not include healthcare related expenses — both of which have increased markedly).

    Worse, the Index has been changed with (“New CPI” formula) so that it is an even more conservative measure for Inflation (as the article link I shared with you and your readers points out). And while I purposely believe that is a conscious intention, one cannot argue the measure was modified and the New CPI measurement provides a more conservative figure for inflation as opposed to the “Old CPI” formula.

    Unfortunately, you seem to be so locked into your published viewpoint that you ignore the very common sense you poffer to others in this Blog.

    It is indeed a disservice to your readers.