Economics 304: Intermediate Macroeconomics
Fall 2007
Reed College -- Professor Jeffrey Parker

Homework #4
Due: 9am, Friday, September 21

Problems from Blanchard: Chapter 6: Problems 1, 5-8.

Empirical Exercise: In this the first part of this exercise you will work with data on unemployment and labor-force participation of men, women, and teens. In the second part, you will download some additional data on unemployment in several metropolitan areas from the Bureau of Labor Statistics Web site (www.bls.gov), import them into an Excel spreadsheet, and perform some numerical and graphical analysis with them.

1. The data set for this problem is located in a spreadsheet HW_4.xls on the server. The data are monthly, seasonally adjusted labor-force participation rates and unemployment rates for the full population over 16, teens (all persons 16-19), men (males 20 and over), and women (females 20 and over).

• Create a time plot (using the line-graph option) with time on the horizontal axis showing the movements in full population, men's, women's, and teens' unemployment rates over your sample period. During what periods have business cycles tended to raise unemployment rates? Have there been changes in the natural rate of unemployment over time? Are your answers different for the three population groups?
• Do the same for the labor-force participation rate. What has tended to happen to the participation rates as the unemployment rate has changed over business cycles and/or due to changes in the natural rate? What trends do you observe over time in the participation rates of the various groups?
• To focus more closely on the differences between the series, use the formula feature of Excel to create new variables for
  • The difference between each group's unemployment rate and the overall rate
  • The difference between each group's participation rate and the overall rate
• Create time plots for these variables.
• What has happened to these differences during cyclical fluctuations in the unemployment rate? What has happened to these differences over time?
• Interpret your results about the behavior of unemployment rates and participation rates of the three groups using standard theories of the labor market and of the unemployment rate.

2. For this problem, you are to identify four metropolitan areas within the United States for study. Download the unemployment rate for each of these cities from www.bls.gov. From the BLS home page, click the link for State and Local Unemployment Rates under the Employment and Unemployment heading, then click Get Detailed LAUS Statistics and Create Customized Tables (Multiple Screens). From the next set of screens, select the states in which your metro areas are located (hold down the Ctrl or Cmnd key to make multiple selections), select Metropolitan Areas on the second screen, select your metro areas on the third screen, then select Unemployment Rate and finally Not Seasonally Adjusted (the only option available). You should now be at a download screen where you can select the options for your download: All Years, Column, No HTML Tables, and Comma delimiter will work best.

Once you click Retrieve Data, you should get your data in long columns one after the other. You can select this long column of data and copy it to the clipboard, then paste it into an empty column in Excel. To align the data properly across columns, keep it selected and go to the Data menu and select Text to Columns. You have commas delimiting your data, unless you chose something else. This should put the data into nice columns. All that remains to be done is lots of copying and pasting to get the variables into columns beside each other with a single set of dates--like the data set in the previous problem. The only problem remaining should be the presence of that awkward Month #13 in the data. These are the annual averages and you want to get rid of them. The quickest way is probably just to delete those rows from the matrix.

Once you get the metro unemployment rates into a nice matrix with only monthly data, copy the appropriate months of data for the aggregate unemployment rate from HW_4.xls alongside the metro rates. Be sure that the months line up properly. Now you're ready to analyze the data.

• Compare the average levels of unemployment in the four metropolitan areas to the national average. Which cities have the highest and lowest unemployment rates? What economic factors would help explain these differences?
• To the extent that your sample period allows you to observe business-cycle fluctuations, in which cities does unemployment seem to be more and less cyclically sensitive? What economic factors would help explain this?