Figure 4. Effects of housing starts on observed labor productivity in single-family and multifamily new housing construction, 1987–2016 (base year 1987 = 100)
Year	Single-family new housing construction	Multifamily new housing construction
1987	18.47	6.55
1988	18.03	6.25
1989	17.48	6.09
1990	16.67	5.68
1991	16.25	4.82
1992	17.67	4.79
1993	18.34	4.72
1994	18.82	5.44
1995	18.00	5.56
1996	18.57	5.79
1997	18.39	5.92
1998	19.28	5.95
1999	19.48	5.91
2000	19.03	5.91
2001	19.29	5.86
2002	19.82	5.95
2003	20.64	5.96
2004	21.27	5.94
2005	21.83	5.98
2006	20.45	5.89
2007	17.79	5.75
2008	14.34	5.60
2009	12.49	4.18
2010	12.79	4.26
2011	12.32	4.86
2012	13.48	5.36
2013	14.30	5.74
2014	14.59	6.00
2015	15.19	6.20
2016	15.76	6.18
Notes: The annual impact of housing starts (hs) is calculated as exp(0.414 · log(hs)) in single-family housing and as exp(0.305 · log(hs)) in multifamily housing. Recession periods are July 1990 to March 1991, March 2001 to November 2001, and December 2007 to June 2009, as determined by the National Bureau of Economic Research. Source: U.S. Bureau of Labor Statistics.