A Real Measure of Inflation

As most of us are aware, the Federal Reserve dreads the prospects of both deflation and inflation.  The Fed has determined that inflation at the rate of 2 percent, as measured using the annual change in the price index for personal consumption expenditures, is most consistent with the Fed's goals of maintaining both stable prices and maximizing employment.  When inflation goes above the target rate, the Fed gets concerned about the long-term economic health of the nation.  When inflation goes below the target (or becomes deflation), consumers may find that their wages are dropping along with prices, causing them to postpone purchases which results in an economic slowdown.

Rather than using the normal Consumer Price Index or CPI, researchers at the American Institute for Economic Research (AIER) have developed a measure of inflation that they call the Everyday Price Index or EPI.  The EPI measures the day-to-day changes in the price of goods and services that people purchase most frequently; these goods and services are purchased at least once every month.   These items include food, utilities, gasoline, personal care products, motor vehicle insurance, various recreational expenses including cable fees and club dues, communication costs and costs of prescription drugs among others.  Unlike the CPI, the EPI deliberately excludes infrequently purchased good including automobiles, computers, electronic goods and payments that are fixed in size by contractual arrangements like rent and mortgages.  Each of the components in the EPI is then weighted by the share of expenditures that it consumes which are equal to the weights used in the Bureau of Labor Statistics CPI calculation.  The EPI is also not seasonally adjusted, rather, it reflects the actual prices paid by consumers.  Most importantly, as you'll note, the EPI assumes that a larger share of household budgets are consumed by energy costs, something that most of us can attest to.  In a world where energy prices are extremely volatile, particularly to the upside, it is quite obvious that increasing costs of energy can have a detrimental impact on household budgeting.

Here is a chart showing the expenditure categories and weights used in calculating the EPI:

As we can see from this chart which shows the increases in prices for each of the components since 2000, the motor fuel component (in black) is by far the most volatile:

Other than motor fuel, the greatest price increases over the decade and a half have been seen in prescription drugs and medical supplies which has increased by 60 percent, auto insurance which has increased by 68 percent, cable and satellite television services which has increased by 58 percent and by fuels and utilities which have increased by a rather hefty 80 percent as shown on this graph:

As shown on this graph, the lowest price increases over the decade and a half have been seen in food at home which has only increased by 42 percent, food away from home which has increased by 48 percent, alcoholic beverages which have increased by 37 percent and club memberships which have increased by 21 percent as shown on this graph:

Let's look at a graph that shows how the CPI (in red) and EPI (in blue) compare over the period from 1987 to the present:

You'll notice that, over the long-term, the EPI suggests that inflation has been far worse in the "real consumer world" than the CPI would suggest.  This is not terribly surprising, particularly given that the CPI includes only goods and services that are rarely consumed.  I think that most consumers would agree that inflation has been far worse than what the headline numbers would suggest. 

In the month of September 2014, the EPI actually decreased by 0.2 percent compared to a 0.1 percent increase in the CPI.  This is because, for the third straight month, the price of gasoline dropped.  The price of home heating oil and electricity also dropped resulting in a total decrease in energy prices of 1.5 percent.   As a result of increases in food prices in September, much of the decline in energy prices were offset.  In particular, the prices of meats rose because of the drought and a deadly swine virus.  

On a year-over-year basis, the EPI increased by 1.6 percent compared to a 1.7 percent increase in the CPI, both of which are below the Federal Reserve's target.  Over the past year, energy prices have been mixed with gasoline falling by 3.6 percent, electricity rising by 2.8 percent and natural gas increasing by 5.8 percent.  On the food side of the equation, the price of pork has risen by 20 percent, the price of beef has risen by 17.8 percent and the price of butter has risen by 23 percent.

It is interesting to see that both methods for calculating consumer price inflation suggest that inflation is coming in below the Fed's target rate.  While we are not in a deflationary environment, if fuel prices continue to drop, the Fed may have to resurrect some form of its monetary policy experiment to counteract a central banker's greatest fears.

The Everyday Price Index - A Superior Measurement of Inflation

Every month in Canada and the United States, Statistics Canada and the Bureau of Labor Statistics    (BLS) release their latest versions of the changes to the prices paid for a basket of goods and services by consumers.  Inevitably, most of us who live in the real world find that the Consumer Price Index or CPI does not reflect the reality of our lives; the data always seems to show a far lower rate of overall price increases than what we find when we head to our local grocery store to buy that roll of ultra-soft toilet paper or bag of potato chips.  This is largely because the divergence in the CPI and the price that we pay has increased, in large part because of 21st century changes in both technology and globalization have affected the prices of large-ticket items that are included in the CPI.  Fortunately, the economists at the American Institute for Economic Research (AIER) have developed the Everyday Price Index or EPI to address this consumer experience.

The Bureau of Labor Statistics or BLS has made adjustments to the components of the basket of goods and services that compose the CPI as they observe that consumers are changing their behaviors.  For example, in the 1980s, Americans and Canadians generally became more health conscious and started to eat more fish and chicken at the expense of beef and pork.  The current list of consumer goods that comprise the CPI includes all types of gasoline, white bread, ground beef, whole chicken, eggs, milk, red delicious apples, navel oranges, bananas, tomatoes, frozen orange juice and ground, roast coffee among other items.  For example, here is a chart showing the price of one dozen large Grade A eggs from 2002 to the present:

Here is a screen capture showing most of the food items captured in the monthly survey by the BLS:

Additional items surveyed by the BLS include the cost of housing, fuel and utilities, household furnishings, appliances, tools, apparel, transportation including public transportation, leasing costs, fuel costs, cost of vehicular accessories, cost of medical care, cost of recreational activities (which includes televisions and audio equipment), cost of education and communication (which includes the cost of computers and software) and the cost of other items including tobacco, personal care products, funeral expenses.  While the list looks exhaustive, as I noted above, the CPI does not seem to reflect the real world no matter how often the BLS adjusts its parameters.

In contrast, here is a chart showing the items included in the Everyday Price Index and their annual increase from 2010 to 2011:

Consumers, when faced with prices for a given item that are "uncomfortably high" will change their behaviors and buy less of that particular item and eventually, they may substitute cheaper goods for the more expensive ones.  While these are rather small effects over the short-term, they certainly add up over the long-term.  Economists at AIER have adjusted the Everyday Price Index (EPI) to allow for these changes in consumer preferences by adjusting the weights of each good or service depending on the proportion of consumer's monthly total spend on each.

Why has the government’s CPI become a relatively meaningless statistic?  Changes in technology have impacted consumers over the past 3 decades; think about your life in the 1980s, you most likely did not have a computer so there was no need for internet access, you had no cell phone, your household probably had only one television and you were hostage to extremely high by-the-minute long distance telephone rates.  Over the decades, these goods and services were either added to or changed to different categories within the CPI.  The inclusion of these items, particularly electronic goods, within the sample used by the CPI has had an unintended consequence; as technology changed rapidly in the last decade, the price for many of these goods has dropped markedly, particularly for computers, cell phones and audio-video equipment as I note below.  Improvements in technology have also led to price declines or price stabilization in items that consumers want (as opposed to what they really need and yes, most of us need these items, we just don't need to purchase them very often) including automobile and household appliance prices; think of what options you get as standard equipment in cars today compared to a decade ago.  On top of all of these issues, globalization has led to a drop in the price of many clothing items and other imported goods, many of which are not everyday purchases.  Since these prices are all included within the CPI, the headline inflation number looks far lower than what we experience since we tend to make more frequent purchases of "needs" like food, fuel and housing where prices are more volatile to the upside than we do of "wants" like electronic equipment and vehicles.  As noted above, the Everyday Price Index excludes these "wants", eliminating the price-reducing forces of both globalization and technological improvements.

As we all know, the price of energy has risen markedly and become much more volatile over the past decade as demand has risen.  The EPI includes energy prices in its calculation, however, since it does not include many of the aforementioned consumer purchases, energy costs make up a larger portion of the EPI than they do in the CPI where their influence on price changes is muted.

Now, let's look at a couple of graphs.  From FRED, here is a graph showing the CPI since 1947:

Here is a graph showing the CPI and changes in prices of selected components of the CPI between 1978 and 2010 noting the relatively steady prices in clothing and new cars over the last two decades as noted above:

It's interesting to note that between 2000 and 2011, that the price of personal computers dropped by 68.7 percent, televisions dropped by 88.7 percent and toys dropped by 49.3 percent.  There goes inflation!

Here is a screen capture showing a comparison between the CPI and the EPI, noting that the two measures of price changes diverge in the early 2000s:

Over the 25 year period from 1987 to December 2011, the average annual CPI resulted in an inflation rate of 2.9 percent compared to 3.6 percent for the EPI, most likely related to the above-noted impacts related to changes in technology and globalization which kicked in around the beginning of the new millennium.

Here is a graph showing how prices increased for various items within the EPI over the 25 year period:

The largest price increases occurred for tobacco products but since tobacco purchases make up a relatively small part of consumer purchases, this price increase did not have much of an impact on the EPI.  The second largest increase occurred in the price of motor fuel and transportation which more than tripled over the 25 year period.  Since consumers spend between 14 and 20 percent of their everyday expenditures on transportation, this category has a far greater weighting in the EPI, as it should.  This factor alone explains at least a portion of the diversion between the CPI and the EPI.

In closing, let's look again at last year.  The BLS's CPI measure showed inflation of 3.1 percent in 2011, up from 1.5 percent in 2010.  In contrast, the EPI showed that Americans experienced price inflation of 7.2 percent in 2011.  Most Americans (and Canadians who, unfortunately do not have a similar mechanism for measuring true inflation) would most likely agree that the rate of inflation established using the EPI is far closer to real-life experience than the measure used by the Bureau of Labor Statistics.

While many of us have suspected for years that government inflation statistics vastly understate price increases in the economy, economists at AIER have now provided consumers with a viable alternative measure of inflation, one that better reflects our personal shopping experiences.