by Calculated Risk on 7/15/2018 08:19:00 AM

According to NBER, the four longest expansions in U.S. history are:

1) From a trough in March 1991 to a peak in March 2001 (120 months).

2) From a trough in June 2009 to today, July 2018 (109 months and counting).

3) From a trough in February 1961 to a peak in December 1969 (106 months).

4) From a trough in November 1982 to a peak in July 1990 (92 months).

So the current expansion is the second longest, and it seems very likely that the current expansion will surpass the '90s expansion in the Summer of 2019.

As I noted last year in Is a Recession Imminent? (one of the five questions I'm frequently asked)

Expansions don't die of old age! There is a very good chance this will become the longest expansion in history.

A key reason the current expansion has been so long is that housing didn't contribute for the first few years of the expansion.  Also the housing recovery was sluggish for a few more years after the bottom in 2011.  This was because of the huge overhang of foreclosed properties coming on the market. Single family housing starts and new home sales both bottomed in 2011 - so this is just the seventh year of expansion - and I expect further increases in starts and sales over the next couple of years.

Unfortunately we only have new home sales data back to 1963, and the housing cycles aren't as clear as the overall business cycle (dates are approximate).  But here are the housing expansions since 1963:

1) From a trough in January 1991 to a peak in November 1998 (94 months).  Note: This could be considered all the way until July 2005 (a total of 175 months!)

2) From a trough in February 2011 to today, July 2018 (89 months and counting).

3) From a trough in June 2000 to a peak in July 2005 (61 months).

4) From a trough in September 1981 to a peak in March 1986 (54 months).

5) From a trough in January 1975 to a peak in October 1978 (45 months).

6) From a trough in February 1970 to a peak in October 1972 (32 months).

Usually housing bottoms a few months before the economy bottoms, but peaks a year or more before the economy peaks (this is why housing is a good leading indicator).   In general the housing cycle is shorter than the business cycle.   By the end of this year, the current housing expansion will be the longest since at least 1963 (unless we consider the previous two cycles as just one - then there is a long way to go!).  

Given the number of housing starts and new home sales, there is still room for further expansion.