Inflation/Deflation: The Devil in the Details


By Kash Hashemi, CFA, MBA

IN THIS ISSUE: How does the government measure inflation? Well what if they decide your brand new $1000 computer is really only worth $500? And they also assume you started buying frozen yogurt instead of ice cream, because ice cream got too expensive? And what if they offset the rise in real estate prices by taking into account lower rent? Inflation can often seem like a vague economic concept, but Portfolio Manager, Kash Hashemi, takes a closer look to reveal the devil in the details.

It seems we can’t get through a single week without hearing someone’s opinion on inflation. We are constantly bombarded by opinions dressed up as facts. The idea of inflation and deflation is a dense and complicated topic, but our challenge is to help you make sense of it all because, to be honest, not all of it makes sense. In this newsletter, we will not focus our attention on recent monthly or quarterly inflation data in an attempt to forecast market outcomes.

Instead, we flesh out the details of the most commonly used and perhaps most misleading measures of inflation: the Consumer Price Index (CPI).

To the average individual, inflation means prices of every day necessities are steadily increasing at a faster rate than their wage is increasing. Instead of being able to walk out of the grocery store having paid $100, he may find he has to pay $110 this year for the same basket of groceries. He realizes he is not making 10% more money this year, so his purchasing power has been compromised and, as a result, his standard of living has decreased.

More often than not, this is the perception most people have about inflation when they read in the papers that inflation is a concern – or could become a concern – because the Consumer Price Index (CPI) data is showing an up tick.

The common theme we’ve noticed amongst the financial media is the idea that inflation and deflation are mutually exclusive. We often hear reporters ask economists whether they are in the inflation or deflation camp as if it has to be one or the other. Of course what these reporters are really asking is whether the data, on average, is pointing towards an inflationary future or a deflationary one – a fair question, but probably not one that can be answered by extrapolating a few months’ data into the future.

What is important for you to realize – not as an investor, but as a working and tax-paying individual – is that the CPI is irrelevant to those concerned about maintaining their purchasing power over inflationary and/or deflationary periods.

Let’s start by providing a basic methodology showing how one can calculate their own personal rate of inflation.

(We will use U.S. data instead of Canadian data, because it is more easily accessible and better researched. We don’t believe this affects the concepts, analysis, and conclusions we put forth.)


We refer to the U.S. Department of Labour for their take on the CPI. They claim the CPI is a statistical average, and may not accurately reflect the personal experience of all families or individuals, particularly those whose spending habits differ substantially from the “average” consumer. Nothing earth shattering here. But the implication is that each individual or family will have their own personal inflation rate, as the CPI will contain items that are a part of some individuals’ buying patterns, but not others’.

The CPI divides the consumer market into 8 major groups and services: food & beverages, housing, apparel, transportation, medical care, recreation, education & communication, and other goods and services. One can estimate the rate of inflation that is applicable to their own situation by giving a weight to each of the 8 groups, calculating the price increase or decrease of each group over a given year, and calculating inflation based on the new relative weights.

Let’s create a simple analogy and look at the example of Joe and Mary Smith, a young couple who’ve bought a home and are starting out. If they were to break down their own personal CPI, it might look something like the above chart. In this example, Joe and Mary’s inflation rate was 4.85% for the year.

There are two ways in which one’s inflation rate can be impacted:

  1. The percentage of the individual’s total spending allocated to each category (column B). This will be very different depending on the individual. For example, an elderly individual is likely to allocate a much higher portion of their total spending on medical expenses than a 20-year-old who would allocate more toward apparel.
  2. The amount one chooses to spend on specific items within a category (column C). Within the categories, there will be different price movements for various goods and services. For example, within the food and beverages category, if Joe and Mary enjoy eating at restaurants regularly, they would likely experience a different inflation rate in that category than the single guy who eats canned tuna every day.

Now that we have a basic understanding of inflation and how it can apply to us, let’s take a closer look at three components that influence the U.S. CPI.

Housing

An important category that often skews the CPI, is housing. This happens because the calculation for the housing category is based on imputed rents as opposed to higher home prices. This estimate is called Owners’ Equivalent Rent and is used as a proxy for inflation in the housing category. We feel this approach results in misleading inflation numbers for the housing component over the short-term, but not necessarily in the long-term.

According to John Mauldin, author of Bull’s Eye Investing, “if you look at the graph of home ownership cost, you find that the numbers are actually very volatile… if you look at a graph of owners’ equivalent rent you find that the volatility is much less and the moves take a longer time… if you put these charts together, it almost looks like the imputed rent is an average mean of the actual costs”1

Mr. Mauldin is saying that while there are times when imputed rents understate actual house price inflation (when housing prices rise rapidly), there are other times when it overstates it (when housing prices are in decline).

During the recent U.S. housing boom, we witnessed many renters become owners. Consequently, we saw house prices inflating while imputed rent prices were flat or deflating as the basic forces of supply and demand took over. Going forward, we will likely see the opposite effect as marginal homeowners become renters again, thereby deflating house prices and inflating rent prices. Such an outcome would result in a positive inflation rate for the housing category of the CPI at a time when house prices are actually falling rather quickly.

What really matters to you, however, is how all of this gobbledygook impacts your personal inflation rate.

Well, if you own a house free and clear, your situation is rather straightforward. It should not be relevant what the market thinks your house is worth, because you are living there and, likely, are not planning to sell and move. An increase in your home price is only relevant for the inflation calculation because it could potentially translate into higher property taxes.

So if you are one of these “free and clear” individuals, the percentage increase in your property taxes (maintenance fee if you are a condo owner and annual home maintenance services if you are a home owner) is the rate of inflation you should use in your personal inflation rate calculation. The reality is that property taxes and home maintenance2 won’t make up a large portion of your total expenditures, so the weight given to the housing component will not be all that high.

The end result for these individuals during the most recent Western Canada housing boom may be a high percentage increase in the inflation rate for the housing component, but a relatively low weight towards this category, creating a muted effect on your overall personal inflation rate.

If you are a home owner with a fixed rate mortgage, you have to make a monthly mortgage payment on top of property taxes and home maintenance. You will likely give a higher weight to the housing component than the individual who owns their home free and clear. But the inflation rate would probably be lower than that of the free and clear home owner because the interest rate on your mortgage is fixed (i.e., 0% inflation).

If you are a renter, on the other hand, the housing component of your personal inflation rate depends on the percentage increase in your rent as well as how much of your total budget you spend on rent.

For renters who are looking to purchase a house or a condo, your “housing” inflation rate will be very high given the recent housing boom in Western Canada. This is definitely not good news since it means you have to shell out a lot more today than 5 years ago to buy the same house or condo.

For these individuals, our recommendation in the past has been to be patient and wait for a more appropriate entry point. This is a challenging task for many since human nature has a nasty habit of equating home ownership with financial security. We just don’t see it that way.


There is also the issue of certain adjustments that are made to the prices of the goods and services in the CPI basket. These are referred to by the Bureau of Labour Statistics (BLS) as hedonic adjustments. Hedonic adjustments occur when this year’s widget is seen as providing more utility to the buyer than last year’s widget.

For example, this year’s computer may be faster than last year’s computer, so the argument is made that it provides a higher level of utility to the user. The price is adjusted downwards to reflect this increased utility and the lower price is used to calculate the increase or decrease in the CPI. So if this year’s computer costs $500, but is twice as fast as last year’s computer, the price will be hedonically adjusted down to $250 for the purpose of calculating the CPI.

This results in understating actual consumer price inflation for products that are constantly being enhanced through new technology.

There are a couple of basic problems in using the hedonically adjusted prices that are embedded in the CPI calculation:

  1. Measuring the extra utility or efficiency gains provided by the new computer is subjective.
  2. Not everyone is going to appreciate the increased utility provided by the new computer.

“In 1998, the methodology was adopted for computers – surely the biggest step backward in realistic inflation calculations. Since then, the BLS has expanded the concept to include audio equipment, video equipment, washers/dryers, DVDs, refrigerators and, of all things, college textbooks! Today, no less than 46% of the weight of the U.S. CPI comes from products subject to hedonic adjustments.”3

Wow!!! It’s not often we use three exclamation marks, but this one really deserves it. 46% is a lot higher than we would have imagined and that was in 2004. We’re still trying to figure out how college textbooks today can provide extra utility than those published in 1998. Are the words and concepts somehow easier to read and understand, allowing students to learn more in the same amount of time? We seriously doubt it, and we fully concur with Mr. Gross. To us, these hedonic adjustments are the height of idiocy.

Substitution

The substitution methodology allows the BLS to replace goods in the CPI basket that have seen price increases with cheaper items that, to the average consumer, could be seen as substitutes. Their argument is as follows: if the price of a particular product increases, consumers will no longer purchase that product and will instead purchase a substitute product that hasn’t seen a price increase.

The BLS uses ice cream as their example. They argue that if the price of a certain brand of ice cream increases, a consumer could buy frozen yogurt instead, or they could substitute their craving for ice cream with an alternative dessert item such as cupcakes.

While we have no doubt that some consumers would react this way, we can’t look past the fact that the actual price increase in ice cream (for example) is not being reflected in the CPI. In other words, the CPI is already assuming that consumers will opt for the cheaper substitute due to inflation and reduced purchasing power.

This, of course, is backwards logic, because the CPI is supposed to be measuring those increases for us so we can make that choice for ourselves.

It turns out these substitutions are being applied fairly liberally to various components of the CPI. As of 2004, substitutions were being used for more than 61% of the total CPI.4

We were admittedly shocked to learn about this method. Simultaneously, it is exactly the type of thing that adds more credence to our argument that the CPI is pretty much irrelevant when it comes to estimating personal inflation rates. This is regrettable because, as you know by now, the C in CPI is supposed to stand for Consumer.

So if it’s not relevant for the consumer, then exactly what purpose does it serve? Unfortunately, the answer is beyond us.

Let’s conclude our analysis with what we feel are the largest inflationary or deflationary forces that clearly impact our lifestyle.

  • The Outsourcing of America (good deflation)
  • Financial Institutions’ lending practices (bad inflation/bad deflation)

The Outsourcing of America

We turn to Richard Russell for his comments on the subject: “The world is now facing massive overproduction, due to the rise of Asian manufacturing. Walk into any major store from Wal-Mart to Macy’s. You’ll be met with such an over-abundance of almost any product you can imagine that the reality is stupefying. The whole developing world is pumping out merchandise at an almost indecent rate. And this over-abundance is deflationary.”5

It is deflationary for two reasons:

  1. Significantly cheaper labour in developing nations means cheaper production costs, which translates into lower consumer prices without sacrificing corporate profit margins.
  2. The supply of these manufactured goods increases while the demand may not increase as much, or at all.

So we can thank the outsourcing of America for the benign growth in consumer goods prices. Without this, we doubt consumer goods prices could have stayed so low for so long.


During the most recent US housing boom, lenders were willing to lend to almost anyone. As long as you had a pulse, you qualified. Non-traditional mortgages flourished from 2000 to 2005.

  • Adjustable rate mortgages
  • Zero down payments
  • Interest only mortgages, and ‘stated income’ applications (where the borrower describes his own financial situation)

In 2000, sub-prime borrowers made up only 5% of mortgages. By 2005, this percentage was up to 25%. Lenders were no longer considering the quality of the borrower or the value of the asset underlying the mortgage. They were merely trying to keep up with the Joneses of the mortgage lending industry. At a time when credit was cheap, speculation was rampant, and most were moonlighting as house flippers, banks and other mortgage lenders made credit easy to obtain. This had a huge inflationary impact in asset markets.

The Mortgage Bankers Association in the U.S. reports that subprime adjustable rate mortgage delinquencies reached 14.4% during the fourth quarter of 2006. Lenders clearly had not considered the consequences of their loose lending practices. We would expect the next chapter of this story to be all about tighter lending practices, where lenders actually start paying attention to the borrower’s ability to repay. Just as loose lending practices result in inflation, tight lending practices result in deflation.

And just to make things more complicated, the type of inflation or deflation caused by loose or tight lending practices do not typically impact the CPI. For starters, home equity extraction and the speculative home buying that followed loose lending practices did not make a bit of difference to the CPI because of the imputed rent calculation we discussed previously. This excess liquidity was used to buy and inflate prices of financial assets, not, for example, underwear sold at Wal-Mart.

After all, if you give a wealthy person easier access to capital, they probably won’t have an urge to drive to Wal-Mart and buy more underwear. And if you give a not-so-wealthy person easy access to capital, chances are they’ll go out and buy a house which won’t get captured in the inflation data anyway.


In summary, the CPI does a very poor job of estimating actual inflation and should not be the focus of consumers when trying to assess how much of an impact price increases have on their spending patterns. Categories such as housing do not represent reality over the short term, especially during periods of significant increase or decrease in general house prices. Add to this the combination of bizarre hedonic adjustments and substitutions, and one is left with a CPI figure that regularly replaces fact with fiction.

To make matters worse, significant external factors that go beyond the simple calculation of consumer goods prices could, from time to time, have a material impact on the consumer. In the end, one has to measure their own personal experience and attempt to estimate their own personal rate of inflation based on real facts and real life situations.