By ANDY ANDREWS
Los Alamos World Futures Institute
In part two of this series we finished by looking at the price of gasoline in 1980 and inflating it using an “average” of inflation rates to today and found the answers wrong.
The calculated answers were significantly wrong because the rate of inflation is based on the Consumer Price Index (CPI), not a single item factored in it.
The “average” price of gasoline in 1980 was $1.262 per gallon (stated to three decimal places as it is today). The word, average, is in quotes because the data I found was the average price per month and I “averaged” those prices. If you say, “sloppy analysis,” I will agree with you. But looking at the same data source and noting the “average” price for April 2019 through March 2020, the value is $2.724 per gallon.
In part two I used a gallon price of $1 per gallon (misstated as $1.19) and a 3.13 percent average inflation rate to find a 2020 price of $3.43 per gallon. Obviously, price and rate data affect the outcome of the model.
In the United States, the Bureau of Labor Statistics (BLS) every year calculates the Consumer Price Index (CPI). This value (actually several different values) is a metric based on a model first developed in 1917. As you might expect, it has evolved and changed over the past 103 years and today it is used to calculate the inflation rate on an annual basis.
The calculated inflation rate is then used to adjust things such as Social Security payments. In 2019, Social Security payments went up 1.6 percent. But does this really reflect the increase in costs to the consumer, individually and collectively?
To perform its calculations, the BLS has a model that includes eight primary categories: (1) food; (2) housing; (3) apparel; (4) transportation; (5) medical care; (6) recreation; (7) education and communication; and (8) other goods and services. If you explore just the list of subcategories under food, you find a total of 85 of them at the bottom of the list. If you go to transportation, you find 24 categories at the lowest levels. One of those categories is unleaded regular gasoline. Obviously, gasoline affects in some way the CPI and, hence, inflation. But gasoline itself does not necessarily adjust in price according to inflation.
Models are used to project or estimate the future based on data from the past with speculative assumptions. As an illustration of the possible fallacy of models, consider the following totally wrong modelling. Assume the price of gasoline today is $1 and you are considering buying (investing in) 1 gallon, storing it for 40 years, and then selling it at the market price. Obviously it is impossible to store the gasoline without deterioration, but assume it can be done.
Next, consider inflation and use the 1980 and 2019 average inflation values I found on the web. In 1980, gasoline process went up by 18.9 percent while in 2019, they only went up an average of 7.9 percent. And just to make the example, more outrageous, let us include the change for 2008 of minus 43.1 percent. Using these values to estimate the value of 1 gallon of gasoline in 40 years you get $1,016.89, $20.93, and essentially zero. Clearly making projections demands accurate assumptions as well as proper data upon which to base them, but it also demands the factor of time.
In considering time, is a 40 year projection a reasonable thing to do? Forty years ago there were no smart phones, there was no Internet available to the masses, and mileage of automobiles was much less. In 1980, the population of the United States was 226.5 million and in 2020 the population is about 330 million. How does this affect supply and demand and how does the evolution of technology impact it? We communicate differently, which impacts the education and communication element of the CPI. We have more cars on the road but with better gas mileage and we are moving toward electric powered vehicles. And we have become more dependent on “instantaneous” information. While we are modelling to predict the future, the evolving future changes the foundation of our models.
As stated above, from 1980 to 2020, the U.S. population has grown almost 46 percent. This changes the values of supply and demand because individuals adapt to both change and availability. In the early 1980s, I owned a Commodore 64 computer. I was able to rig it up to control an electronic typewriter that typed carbon ribbon letter that I could mail. Email was not available and, if I could access the electronic network, the best I could expect for data transmission and receipt was 300 bits per second. Today we measure the flow of our data, not all data, just data for us in megabits per second. And one megabit service is probably inadequate for streaming entertainment. Since the beginning of the consumer price index, there have been many changes allowing it to adapt to “modern” times. The model is evolving as humanity evolves. Can it be used to predict distant prices? Probably not, but it does give us short term decision support while we gather more data. Yet it uses large quantities of current data that must be assumed valid in a broader context for the short term decisions. But do the short term decisions affect the data base?
Til next time….
Los Alamos World Futures Institute website is LAWorldFutures.org. Feedback, volunteers and donations (501.c.3) are welcome. Email andy.andrews@laworldfutures.org or email bob.nolen@laworldfutures.org. Previously published columns can be found at www.ladailypost.com or www.laworldfutures.org.