The Distress Index (A Better 'Misery' Index)
I was thinking a few months ago that FEE should develop a new and more accurate “Misery Index”. The original “Misery Index” is really not all that useful since it measures only unemployment and inflation. Deflating prices bring the index down but do not necessarily indicate a healthy economy. Sometimes it’s the opposite. We need something that is more applicable to our current economic situation.So the other day, I emailed the idea to a few of FEE’s friends in the economic world and lo-and-behold, Paul Cwik, an economics professor at Mount Olive College, took the idea and made it real. He revamped my initial suggestion and put together a simple but powerful index of five statistics that seem to give us a pretty good sense of what’s going on in the economy. The whole thing came together in just a few hours and has not yet been properly scrutinized. But the initial results are very exciting.
Methodology
We wanted to keep the index simple, no more than a handful of statistics and it was important that those statistics be relatively uncontroversial. So we relied solely on numbers provided by the Federal Government.Included Statistics:
- Unemployment: Clearly, no “misery” index would be very relevant without considering unemployment. This is pretty self evident.
- Consumer Price Index: Like the original “misery” index, we included inflation, even though we are actually in a deflationary period at the moment.
- Gross Domestic Product: Wikipedia describes GDP as a “basic measure of a country’s economic performance and is the market value of all final goods and services made within the borders of a nation in a year.” Clearly this is an important part of understanding the state of the economy.
- Total Capacity Utilization (TCU): This is a measure of the utilization of the all available capital goods. We use the inverse of this number, since higher utilization is generally a good thing. So for instance, if TCU is at 70 percent, we would add 30 percent to our index as a measure of the idle capacity.
- Household Financial Obligations as a percent of Disposable Personal Income (HFO/DPI): This measure is intended to gauge the ability of individuals to participate in the consumer economy.
It is important to emphasize that no statistic will ever fully articulate what is happening in the real economy. The real economy is made up of living, breathing, planning, acting individuals. Statistics are simply an abstraction and, as such, imperfect. Nevertheless, we feel this index has substantial value for two reasons.First, it gives us a tool to help interpret what the media and government are telling us about the economy. Second, we hope it will give voice to the taxpayer and the frustrating conditions he or she is enduring these days. We hope the index will keep pressure on policy makers and opinion leaders to make decisions that improve the economy rather than distressing it further.
Current Distress Index: 61.0
Unemployment: 9.7%
CPI: -1.5%
GDP: 3.897%
TCU: 30.4%
HFO/DPI: 18.5%
Historical Distress Index
After formulating the index, Professor Cwik ran a cursory historical analysis on the index. The results were pretty impressive. The chart below shows the Distress Index since 1967 with economic recession periods highlighted. There seems to be at least a superficial correlation between the index breaking 45.0 and the economy falling into recession. (Note we have not tested the strength of this correlation). In most cases the index appears to lead the recession’s beginning and end, which would seem to indicate that the index is actually useful in telling us where we are headed, not just where we’ve been.
Distress Index (click to enlarge)
We invite you to criticize and/or improve this index. We eventually plan to post this index at fee.org and publicize it heavily. You can download the data here.
Comment by Fariz on 19 September 2009:
I think you could add government debt/GDP ratio here, because this adds a lot of distress in many countries. Besides, many countries grow by heavy borrowing, we should keep in mind that those who are borrowing less deserve distinction, although borrowing might say how much we trust to those countries soundness. again, i think you should put that item into picture here. Besides, don’t you think capacity utilization somewhat effects unemployment?
Comment by Michael Dahlen on 19 September 2009:
Two criticisms: 1) Wouldn’t you want to subtract GDP instead of add it since higher GDP is a good thing? 2) It might be a good idea to provide weightings for these measures. The TCU figure is practically as high as the other 4 measures combined.
Comment by Paul Cwik on 21 September 2009:
The idea on including a Debt-to-GDP is a good one. We’ll have to see what it does to the graph.We are actually subtracting %-change in GDP from the index. So the more the economy sinks, the larger the misery index.We are taking TCU and subtracting it from 100%. So an 80% capacity is than treated as a 20% idleness. Again the idea is that when there is less economic activity the misery index goes up.
Comment by Paul Cwik on 21 September 2009:
Okay, so I have looked at the Debt/GDP ratio and the number just explodes upward over time. While there is a short period in the late ’90s where it goes down, it isn’t really adding much to the index. It is really just getting worse and worse and not changing as the economy changes.So while this is a problem for freedom lovers, it’s not helping with this particular index.
Comment by Fariz on 21 September 2009:
Paul, thanks for considering my suggestion. You might want to use deviation from long term equilibrium in Debt/GDP ratio using Hoddrick-Prescott filter. That might stop that explosion, I guess.
Another variable like this might be budget deficit/GDP ratio.
Comment by Paul Cwik on 22 September 2009:
Fariz, I looked into the impact of using deviations from the 12-month moving averages for TCU, but the impact wasn’t significant enough to make the change. Additionally we are moving away from the simplicity of the index. The original “Misery Index” was inflation plus unemployment. We are trying to keep simple as well. If this was going to be used as a full blown prediction model, then you’d probably be right, in that deviations from trends are better. However, for what we are doing, we aren’t getting enough benefit for the effort.
Comment by Fariz on 22 September 2009:
Paul, I agree with your stand. Simplicity should be compromised only for substantial improvement.
Comment by Michael Shanklin on 24 September 2009:
Good work Dr. Cwik…..sending out to thousands of people now…..
Comment by Glen Bradley on 25 September 2009:
I think a pretty simple modification could make the index more accurate over time. Rather than a raw GDP, which does not account for population growth etc, use GDP/percapita as the basis to calculate the percentage for the index, and it will stabilize the index over periods of 50 years.
Comment by Paul Cwik on 28 September 2009:
Glenn, I have been looking at using GDP per capita. Here are my problems: I have monthly data for three of the statistics. GDP is reported quarterly so there is a loss of accuracy there. However, the population numbers that come out are annual. So that means I only get one reading per year. (See: http://www.ers.usda.gov/Data/macroeconomics/Data/HistoricalRealPerCapitaIncomeValues.xls) So, while the idea is sound, the data doesn’t seem to exist. Do you know where some more precise data might be? If so, please send it!
Comment by Fariz on 28 September 2009:
Paul, I think NBER and a company called eforecasting used to calculate monthly GDP numbers. They might also have GDP per capita if they keep track of population numbers.
Comment by Paul Cwik on 28 September 2009:
Fariz, The NBER has the following to say,
“We view real GDP as the single best measure of aggregate economic activity. In determining whether a recession has occurred and in identifying the approximate dates of the peak and the trough, we therefore place considerable weight on the estimates of real GDP issued by the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce. The traditional role of the committee is to maintain a monthly chronology, however, and the BEA’s real GDP estimates are only available quarterly.” … “We also look at monthly estimates of real GDP such as those prepared by Macroeconomic Advisers (see http://www.macroadvisers.com).”I then went to macroadvisors and they are a subscription service, which I have not signed up for. Sorry. You don’t happen to know someone who has a subscription?
Comment by Fariz on 28 September 2009:
Hmm. You may want to try this link below. http://www.e-forecasting.com/US_Monthly_GDP.htmlSee if you can replicate their numbers for other pediods.
Comment by Paul Cwik on 28 September 2009:
Fariz, Thanks for the suggestion. I looked at the free data they have and it covers July 2008 – July 2009. Here’s the next problem: I’m currently using % change year-to-year. Now, I cannot really use “% change year-to-year” because they aren’t providing that data, at least not for free. (Or at least that I could find.) So then I looked month-to-month and now I have a scale problem. A month-to-month change is going to be small unless it is annualized as (1+x%)^12. So I did that too and while I get a similar result as before, it has added alot of new “ups and downs.”That being said I could then start to do some smoothing and deviations from some moving averages, etc.. Then I come back to the purpose of this index: to use readily available data that anyone can find. Here we’re getting away from this concept and while the goodness-of-fit improves, I don’t think it adds enough. Of course this is not for a peer reviewed journal nor a statistic that I plan to make millions from playing the market. I’m thinking that this index is more like when a weatherman says “Tomorrow is going to be partly sunny.” Why not “mostly cloudy”? Is there any real difference in my life between the two? So I’m very pleased with the index and I am very impressed with the feedback I have been getting. At this point I am fairly comfortable with it and I think I can recommend it to FEE.
Comment by Fariz on 28 September 2009:
Sure. I don’t see any reason not to recommend it. Good luck! Looking forward to other indices…
Trackback by stop spam plugin on 8 February 2012:
Blog Spammed…
[...]Here is a Great site You Might Find Interesting that we Encourage You To See For Yourself[...]……