Please turn on JavaScript

Brooks Wilson's Economics Blog: The 4 Week Moving Average of Initial Unemployment Claims and the Unemployment Rate

Monday, October 12, 2009

The 4 Week Moving Average of Initial Unemployment Claims and the Unemployment Rate

Robert J. Gordon did research looking at the relationship between the 4 week moving averages of initial unemployment claims and found that recessions often bottom out shortly after the 4-week moving average of initial unemployment claims peaks. Barring a deep double dip recession, the average peaked at 658,750 for the week ended April 4, 2009.  On October 8, 2009 the Department of Labor released the most recent data on initial unemployment claims for the week ended September 19, 2009 in "Unemployment Insurance Weekly Claims Report."  Seasonally adjusted initial claims was 521,000, down 33,000 from a revised estimate of initial claims of 554,000 for the week ended September 26, 2009. The 4 week moving average decreased 9,000 to 539,750. The average is down 119,000 from its peak, signaling a probable peak for this business cycle.

Using National Bureau of Economic Research estimates on the beginning and ending dates of recessions, I have  included graphs that compare the recessions that began in March 2001, July 1990, and July 1981 with the current recession which began in December 2007.  Each compares the 4 week moving average of unemployment claims and the unemployment rate of the recession beginning in 2007 to those of the previous recessions.  I have not attempted to adjust the initial claims data for changes in the size of labor market or the unemployment rate for changes in the natural rate of unemployment. The plots are measured over 108 weeks, beginning eight weeks before the recessions began. The horizontal axis begins in October 2007, the date the current recession began, and the data for the other recessions are superimposed on this date. The graphs gives some insight into why economists, politicians and others have expressed so much concern about the current recession.

The current recession seems to have the depth of the 1981 recession but the 4 week moving average has fallen more slowly now than in the 1981 recession.  Unemployment peaked at 10.8% and began to fall in January of 1983, but by May, the unemployment rate of 10.4% was still higher than the current unemployment rate of 9.8%.

The 1990 recession was shallower than the current recession both in terms of length, and depth.  It lasted eight months from peak to trough; the current recession has not officially ended.  The 4 week moving average hitting 501,250 forty-two weeks after the 1990 recession began compared to 658,750 seventy-four weeks after the current recession began.  The recovery from the 1990 recession was slow.  Unemployment was still rising two years after the recession began and would peak at 7.8% in June 1992, two months later than the time frame pictured in the graph.  Still 7.8% is two percentage points lower than the current recession and we do not know when it will peak.

The recession which began in 2000 was similar in length, and depth to the 1990 recession.  It lasted eight months from peak to trough, with the 4 week moving average topping at 517,00 thirty-four weeks after the recession began.  Unemployment peaked at 6.0% one-hundred weeks after the recession began.  Again, the current recession will last at least as long, experience higher levels of initial unemployment claims and unemployment.       


  1. Unemployment is one of the major is at hand.Our unemployment over the years has really had a bad impact on society and it is getting even worser by the moment.

  2. Michelle Toups2/11/09 7:51 AM

    It is interesting how what happens in the past really does pertain to the future. We are able to go back and look at the past recessions to draw conclusions about the one we are in. Unemployment is a major problem, so, knowing this, we should constantly be looking for the first signs of it. We know what it looks like, we know when it will get bad, we even have it pinned down to when it will likely be over. Now, we must use the information to decrease unemployment and increase economic prosperity in our country.

  3. Tammy Mader3/11/09 9:19 PM

    I agree with michelle. why cant we study the past more to help us better prepare for whats going on now, and for the future. true, we cant be 100 percent correct on the predictions about the recessions in the future, but we can definately be able to recognize the signs of one.

  4. CJP said...

    Unemployment is always going to be a growing factor no matter the time period. When it comes to studying the past..i almost feel like we dont do it quite enough..because if we did we would be completely ready for what we get hit with now a days. If we did all this the right way maybe we can lower the number of unemployed people and get people where they need to be economically!

  5. Kaitlyn Wooley8/11/09 2:27 PM

    I believe that we should be smart enough to study the past so that we do not repeat the mistakes we made previously. If we look at what all happened in the past we can learn from it and maybe not make the same stupid mistakes again. Recession is going to happen. But by learning from the past we can figure out ways to get us out of this current recession faster and in the best way possible.

  6. Laura Ehlers8/11/09 8:19 PM

    Economists should use the experiences from the past to make predictions about current events. This recession is proving to be one that will be memorable. I personally know some discouraged workers and some people who are working at jobs that are below their educational qualifications. Graphs just don't seem to show the true dispair in our country now. Why aren't our leaders willing to make changes to strengthen our economy?

  7. Ken Haltom said...

    While the four week average of initial unemployment claims may be a good indicator of when a recession turns the corner, the unemployment rate is a lagging indicator of an economy's progress from recession to recovery. When a recession continues for many months, long term unemployed workers may become discourage from looking for work and quit thereby not being counted and making the unemployment rate look less severe than it is. Likewise when a recovery is beginning and the discouraged unemployed begin to start looking for work again, the unemployment rate remains high and creates a perception that the recovery is nonexistant or is recovering at a slower rate than it really is recovering.

  8. Ken Haltom said...

    One thing that the unemployment rate does not reflect is what are employees quality of life after obtain new employment. Has there pay been cut by 10% or more? What have they done to readjust? Have they cut out optional spending such as eating out and vacations or have they cut out essential expenses such as health and life insurance premiums, contributing to a retirement plan, or putting student loan payments into deferrment? One would hope they cut out the former and not the latter, otherwise there could be more economic hardship down the road.

  9. Andi Thomas28/3/11 12:59 PM

    It is very interesting to see the correlation between the past and the present. Some could say that this is the worst recession of the past 3 decades. But as in the past, we will pick back up. It may take a long time and it may mean unemployment is high until then but unemployment will eventually go back down and the economy back up.

  10. While our past is our best way to look at the future, I am still weary of the future. Although it is shown that our country has gone through recessions and come back from them doesn't mean we should just assume it will get better.