Inflations

Foreword

For as long as I can remember, inflation measures have been criticized as being inaccurate, biased towards the current political ruling elite. The complaints generally assert that inflation was under reported to serve some nefarious political means, usually related to re-elections or other political goals. At the same time, numerous alternative inflation measures were introduced, almost all with much higher numbers than the official inflation figure, some even consistently in the double digits.

What gives?

As usual, a reminder that I am not a financial professional by training — I am a software engineer by training, and by trade. The following is based on my personal understanding, which is gained through self-study and working in finance for a few years.

If you find anything that you feel is incorrect, please feel free to leave a comment, and discuss your thoughts.

Looking outside the window

As I type this, I look outside the window at what looks to be a dreary day — overcast skies and just generally gloomy. At the same time, Google is reporting that the temperature is 82 degrees Fahrenheit (28 degrees Celsius), a warm, bright sunny day.

As I shake my head at the obvious lies the government is feeding us, trying to convince us of global warming to justify the nonsense Green Transition, I changed the Google query from “weather tokyo” to “weather toronto”. Now it says 58 degrees Fahrenheit (14 degrees Celsius), with a heavy downpour. Well, which is it?! Is it hot and sunny or chilly and raining?!

Looking outside the window that is cloudy but clearly not raining, much less a downpour, I can only sigh. What is the world coming to, when even Google’s weather reports are so influenced by the government that you cannot trust it to tell you the weather, here, in New Jersey?

Headline inflation

As we all know, the preferred inflation measure of the US government is CPI inflation, sometimes called headline inflation. More accurately, this is the Consumer Price Index for All Urban Consumers (CPI-U) inflation. This inflation measure is the rate of change of the CPI-U index, itself a composite measure of what the average urban consumer pays for their goods and services in the measurement period.

To simplify it a lot, the CPI-U index is the “composite price” of a basket of goods over time, and CPI inflation is just the rate of change of that composite price. You can find the basket of goods, their relative weights, and the prices of those goods in data from the Bureau of Labor Statistics (BLS), for example, here.

To recap – CPI inflation is based on a known methodology (rate of change) of the CPI-U index. The CPI-U index is a composite index with publicly released methodology on how it decides what is or is not included, with details on the underlying components, their weights and prices, all freely available.

Conspiracy 1

One of the most common conspiracies about the inflation measure is that it is opaque, and legislators/regulators change it at a whim to sugar coat bad economic numbers or to somehow achieve some political gains.

As we can see from above, that is.. quite a far from the truth. The methodologies and details are all public information, and freely available. Yes, they are provided more than a month after the fact, and the data is often revised. But if you think about the scope of the project, the amount of data that needs to be collected, the number of people that need to be sampled and interviewed to determine the basket, it should be obvious that it’s hard to get everything in place in real time, and often initial estimates will be slightly wrong as more data comes in.

Conspiracy 2

The CPI-U methodology (not the data) is changed intentionally over time to make things look better.

Yes, the CPI-U methodology has changed over time. As research into consumer behaviors increase, and as we understand more about economics and finance, it often becomes clear that previous models are less accurate, and new models are made.

This is a common practice in the hard sciences, for example, we used to believe in Newtonian physics, but Einstein introduced Relativity and Einsteinian physics which are believed to be more accurate on a galactic scale. Despite this, Newtonian physics is still taught in schools, because it is pretty accurate when dealing with terrestrial matters and it is much easier to grasp. I don’t hear anyone screaming that the government is up to some nefarious purpose with regards to these 2 physics, do you?

The fact is that economics is not a hard science — there is generally no real way to conduct experiments to verify hypotheses. As a result, there are a lot of assumptions and unknowns in economics which, over time, get refined. Inflation is an area of economics that is most in debate, because of its significance and because so little is understood about it.

And to make it clear that there doesn’t seem to be some higher conspiracy at work, the BLS typically publish the CPI-U for historical dates using the new methodology when it does change, for example here.

So yes, maybe inflation is slightly better now with the new methodology, but it is also probably better for previous periods, which gives us a frame of reference.

Conspiracy 3

My favorite (not) conspiracy is that the government is intentionally lying to us by publishing only one inflation measure and claiming it is the unblemished truth.

If you pay attention to the news, you’ll see that the news outlets typically equivocate “inflation” to “CPI inflation” as we’ve discussed above. The claim by the conspiracy theorists is that this is intentional! That this one number, that clearly does not reflect anybody’s, certainly not their own, experiences is forced upon us for some ulterior motive.

First of all, this is utter bovine feces. There are many measures of inflation, and anybody bothering to spend 5s on Google can find that the government even explains why there are so many, why it’s so hard, and why certain measures are used, for example, here.

For government purposes, there are 2 main variants of inflation measures — the CPI series and the PCE series. There are various differences between how CPI and PCE collects data which I will not go into, and more importantly, each of these 2 series also have different methodologies applied to them to arrive at different inflation numbers.

For example, if you have paid attention in the past ~4 years, you’ll probably have heard of “trimmed mean” inflation, “sticky” inflation, “median” inflation, etc. These are different methodologies applied on the same underlying data (either CPI or PCE) to arrive at different numbers.

Of these, the trimmed mean inflation has a bad rep because it was accused of being conjured out of thin air to present a better picture during the Covid inflation scare of 2021/2022. The truth is that this measure was first proposed in the early 90’s, and has been published for a long time, with historical data points backfilled using the same methodology.

Trimmed mean inflation is just inflation calculated excluding the outliers. There are various ways to determine which outliers are, and different measures may use slightly different methodologies here, but the basic idea is one rooted in statistics. In statistics, an outlier is a data point more than 2 standard deviation away from the mean. The idea is that by removing these outliers, which are likely due to measurement errors or noise, we can arrive at a cleaner signal. Similarly, the trimmed mean inflation measure is intended to provide a cleaner read of inflation trends, by removing what may be erroneous data points.

So, to recap, there are many, many inflation numbers, for example, the regular CPI inflation, PCE inflation, trimmed mean CPI inflation, trimmed mean PCE inflation, median CPI inflation, median PCE inflation, etc., etc. And the government also publishes other measures of inflation outside of the CPI/PCE series.

And if you don’t like any of these, the good news is that the underlying data is all freely available, so you are free to pick your own basket of goods, their weights, and the methodology to determine your own inflation rate.

Conspiracy 4

There are many inflation measures and the government chooses the one that it likes the most for propaganda purposes!

Believe it or not, this argument was made to me by the same person who made conspiracy theory 3. Yes, go figure.

As mentioned above, to make policy decisions, the government (or its various agencies) needs to pick one inflation measure to use. They may look at all of the inflation measures, but there will always be one that is preferred, due to their particular purposes.

For example, for cost of living adjustments to social security payouts, it makes sense to look at the CPI series of data, because that series reflects prices that consumers pay. Of the CPI series of data, CPI-W (Consumer Price Index for Urban Wage Earners and Clerical Workers) was determined to be most representative of Americans on average, so that’s the one used.

Notice the bolded terms, on average. A single number must be used to adjust for whatever policy is at hand, and clearly the one representing the average American is the one most representatives of all Americans in this case. That doesn’t mean that all Americans will see the same inflation! For obvious reasons, everyone spends their money differently, and so will experience different levels of inflation. But for policy purposes, across all Americans, we can’t just pick the inflation that ConspiracyTheoristA faces and apply that to everyone else, can we?

So yes, the inflation measures that each government agency chooses may not represent you, but that’s more of a feasibility limitation than a conspiratorial one.

Other government agencies prefer other inflation measures because it makes more sense for them. For example, the Federal Reserve prefers “core” inflation measures using the PCE series for policy decisions. Core inflation measures strip out items that are generally not affected by interest rates, which generally means core inflation measures do not include food and energy items.

Yes, it is ironic that the Federal Reserve ignores food and energy, 2 of the largest components of inflation experienced by most everyday Americans, but there is a good reason. The Federal Reserve mainly has control over short term interest rates. However, food prices are mainly tied to weather (and thus how crops perform) and energy prices are heavily tied to geopolitical developments (e.g. instability in the Middle East). As such, it doesn’t make sense for the Federal Reserve to make short term interest rates policies based on food/energy prices, because those things are not very affected by short term interest rates in the first place!

Imperfect

The above argues that inflation measures are probably not nefarious in design, but that doesn’t mean they are perfect. And if your argument is that inflation measures are generally flawed, then I will agree completely with you!

Inflation is a subject that is emotional and very poorly understood. So any model of inflation, and any measure of it, is almost by definition wrong. That said, it should be noted that, all models are wrong, but some models are selectively useful.

The topic of what’s wrong with inflation measures is long, and best left for another day. The point I want to make in this post is that in most cases, it seems like the government is making the best of a bad situation, and adjusting their methods as more is learned through research. Maybe on the margins there are some shenanigans going on, but it doesn’t appear, to me, that there is institutional bad faith in policy decisions.

Average, imperfect world

You probably think I’m an idiot, complaining that the weather report for Tokyo and Toronto does not reflect what I’m personally experiencing here in New Jersey. And you’d be right.

Similarly, it doesn’t make sense to complain that any particular inflation measure doesn’t reflect your personal experience — they aren’t meant to. There are many different measures of inflation, for different purposes. Some are tailored to particular segments of the economy, or to particular regions of the country. Depending on how you spend your money, one or more, or none, of these inflation measures may apply to you.

It is, ultimately, up to you to figure out which inflation measure best reflects your personal situation, just like it is up to me, to figure out which weather measure best reflects my experiences.

Just understand that for policy purposes, one single measure needs to be chosen as representative for inflation. It doesn’t mean that some shady government officials shrouded in shadows are decreeing that measure is what everyone must be experiencing! It’s just that for whatever reasons, that measure is deemed to be most appropriate for the situation.

Personal finance

Quick note: It should be obvious from the above, that using CPI inflation for your personal financial planning is probably not the best idea.

Yes, it may represent the average American pretty well for certain purposes, but it probably doesn’t reflect you.

If you are so inclined, you can use the raw data published by the BLS, and your actual spending, to determine a better inflation measure for yourself.

But this is tedious and hard and easy to get wrong — there are legitimate one time expenses (e.g. roof repairs) that will skew the numbers and make things seem better or worse than they really are. Trying to properly adjust/account for these is a science unto itself, which I will not go into.

Personally, being the lazy slob that I am, I typically just take CPI inflation and add 1% to it. Which is to say, for most intents and purposes, for long term planning, I assume a personal inflation rate of 3% (2% Federal Reserve target +1% headroom).