Wisconsin Unemployment Rate Forecast

 

Bruce E. Hansen

Mary Claire Aschenbrenner Phipps Distinguished Chair

T. Haavelmo Professor

Department of Economics

University of Wisconsin

 

10 August 2017

 

This memo reports a 12-month forecast for the seasonally adjusted Wisconsin unemployment rate. In addition to point forecasts (the expected future value of the unemployment rate), the memo also reports 50% and 80% forecast intervals (probable ranges for future values).

The unemployment rate in May 2017 was 3.1% roughly constant since April, but down significantly from 2016. The unemployment rate fell significantly during 2017.

The forecasts are summarized in Figure 1 and Table 1. The point forecast is for the unemployment rate to remain steady at 3.1% for the remainder of 2017, and then rise during the first half of 2018 to 3.5% by June 2018. The 80% forecast intervals show that there is considerable additional uncertainty. There is a possibility that the unemployment rate could continue to decrease, possibly as low as 2.4% by June 2018. It is also possible that the unemployment rate could substantially increase, to 4.8% by June 2018. The 50% forecast intervals refine this uncertainty, showing that it is unlikely the unemployment rate will decrease below 2.9% over the next year, or increase over 4.0% within the next year. Overall, the forecast is for the unemployment rate to stay roughly constant or potentially increase during 2018.

A 50% forecast interval is designed to contain the future unemployment rate with 50% probability. It is just as likely for the rate to fall in this interval as out of it. This is the smallest possible interval which has even odds of containing the future rate. We can think of this interval as “likely” to contain the future rate.

An 80% forecast interval is designed to contain the future unemployment rate with 80% probability. We can think of this interval as “highly likely” to contain the future rate. The 80% interval is designed so that there is a 10% chance that the future value will be smaller than the forecast interval, and a 10% chance that the future value will be larger than the forecast interval.

To understand the economic reason behind these forecasts, the econometric model finds the following salient features. The state unemployment rate is below its long-term average. Mean reversion predicts an increase of about 0.50. Similarly, the U.S. national unemployment rate is low, accounting for an increase of about 0.20 over the upcoming year. The spread of low-grade corporate bond yields over investment grade is lower than average, accounting for a predicted decrease of about 0.3.  Housing starts are below their long-term average, accounting for a predicted decrease of 0.6. Building permits are also below their long-term average, accounting for an increase of about 0.5 over the upcoming year. The other variables contribute only small effects to the forecast. Together, most of these effects offset with the net effect of a short-term decrease followed by a modest increase.

 

Figure 1: Wisconsin Unemployment Rate Forecasts


 

TABLE 1: Wisconsin Unemployment Rate Forecasts

 

History

Point Forecast

50% Interval Forecast

80% Interval Forecast

2017:1

3.9%

 

 

 

2017:2

3.7%

 

 

 

2017:3

3.4%

 

 

 

2017:4

3.2%

 

 

 

2017:5

3.1%

 

 

 

2017:6

3.1%

 

 

 

2017:7

 

3.1%

(3.0%,  3.1%)

(3.0%,  3.1%)

2017:8

 

3.1%

(3.0%,  3.2%)

(3.0%,  3.2%)

2017:9

 

3.1%

(3.0%,  3.2%)

(2.9%,  3.3%)

2017:10

 

3.1%

(3.0%,  3.3%)

(2.8%,  3.4%)

2017:11

 

3.1%

(2.9%,  3.3%)

(2.7%,  3.5%)

2017:12

 

3.2%

(2.9%,  3.4%)

(2.7%,  3.6%)

2018:1

 

3.2%

(2.9%,  3.5%)

(2.7%,  3.8%)

2018:2

 

3.3%

(2.9%,  3.6%)

(2.6%,  3.9%)

2018:3

 

3.3%

(2.9%,  3.6%)

(2.6%,  4.1%)

2018:4

 

3.4%

(2.9%,  3.7%)

(2.6%,  4.2%)

2018:5

 

3.4%

(2.9%,  3.8%)

(2.5%,  4.5%)

2018:6

 

3.5%

(2.9%,  4.0%)

(2.4%,  4.8%)

 

Previous Forecasts                

                June 2017

April 2017

February 2017

January 2017

December 2016

November 2016

October 2016

August 2016

July 2016

June 2016

May 2016

April 2016

March 2016

February 2016

January 2016

December 2015

November 2015

October 2015

September 2015

August 2015

July 2015

June 2015

May 2015

April 2015

March 2015

February 2015

January 2015

December 2014

November 2014

October 2014

September 2014

August 2014

                May 2014

                April 2014

February 2014

January 2014

December 2013

September 2013

                August 2013

July 2013

June 2013

May 2013

April 2013

March 2013

February 2013

January 2013

December 2012

November 2012

October 2012

September 2012

August 2012

July 2012

June 2012

May 2012

                April 2012

Februrary 2012

January 2012

December 2011

November 2011

October 2011

September 2011

August 2011

July 2011

June 2011

May 2011

                April 2011

March 2011

February 2011

January 2011

                December 2010

November 2010

October 2010

September 2010

August 2010

July 2010

June 2010

May 2010

April 2010

                March 2010

                February 2010

January 2010

                December 2009

                November 2009

 

Forecast Methodology