Homebuying Demand Still Strong in Job-Creating Affordable States According to NAR September 2018 Survey

 In affordability, Buyer Traffic, Economist Commentaries, Home Price, Mortgage, REALTOR® Confidence Index, state data

Based on REALTORS® who reported about their transactions in September 2018, home buying activity slowed compared to the level of activity one year ago on a national level, according to the  September 2018 REALTORS® Confidence Index Survey. The REALTORS® Buyer Traffic Index[1] slid to 51 in September 2018, ten points lower from one year ago to (57 in August 2018; 61 in September 2017). Homebuying is now hovering in the “stable” mode, with the index hovering around 50. The REALTORS® Buyer Traffic Index is a leading indicator of closed sales, with a lead of one to two months.

Homebuying demand has slowed overall, but in states with strong job growth and affordable house prices, REALTORS® reported that homebuying demand remains “strong” compared to one year ago.

REALTORS® reported “very strong” (75+) buyer traffic in Idaho and “strong” buyer traffic (55+ to 75) in 35 states such as Indiana, Wyoming, Utah, South Carolina, Michigan, Pennsylvania, Kentucky, Ohio, Missouri, North Carolina, Nebraska, Arizona, Nevada, Georgia, Tennessee, New York, Massachusetts, Pennsylvania, Maryland, and Virginia.

On the other hand, buyer traffic conditions were essentially “stable” (45+ to 55) during July-September 2018 compared to conditions one year ago in states such as Connecticut, California, Oregon, Colorado, Washington, North Dakota, North Dakota, Iowa, Oklahoma, Texas, Florida, Hawaii, New Jersey, and the District of Columbia. Many of these states had “strong” to “very strong” buyer traffic conditions (55+) in last year’s surveys for these same reference periods. The combination of rising interest rates and already high prices (perhaps along with the expectation that prices may fall) may be causing buyers to step aside for now.

Homebuying demand is still strong in states with strong job growth and affordable home prices. I calculated the share of mortgage payment (principal plus interest) to the estimated household income[2] in 2018 Q2 for 160 metro areas for which NAR reports the median sales prices. The graph below shows that home prices are still affordable in 133 metro area, where households typically spent no more than 25 percent of household income on mortgage payment in 2018 Q2.[3]

Among metro areas that generated at least 10,000 total non-farm payroll jobs from 2017 Q2 through 2018 Q2[4], the most affordable metro areas were Saint Louis, MO (15% of income spent  on mortgage payment); Cleveland-Elyria, OH (15%); Oklahoma City (16%); Tulsa, OK (17%);Grand Rapids-Wyoming (17%); Columbus, OH (18%); Indianapolis-Carmel-Anderson, IN (18%); Kansas City, MO (18%); Philadelphia-Camden-Wilmington (18%); Memphis-TN (19%);Baltimore-Columbia-Townsend, MD (20%);and Minneapolis-St. Paul-Bloomington, MN-WI (20%). Dallas-Fort Worth-Arlington, TX stands out for creating 118,000 jobs while having an affordable housing market, with households spending 22 percent income on mortgage payment.

Among metro areas that generated at least 10,000 total non-farm payroll jobs from 2017 Q2 through 2018 Q2, the most unaffordable metro areas were San-Jose-Sunnyvale-Sta. Clara, CA (66% of income spent on mortgage payment); San Francisco-Oakland-Hayward, CA (54%); San Diego-Carlsbad, CA (47%); Miami-Fort Lauderdale, FL (35%); Seattle-Tacoma-Bellevue, WA  (33%);Denver-Aurora-Lakewood (32%);Riverside-San Bernardino-Ontario, CA (31%);and Portland-Vancouver-Hillsboro, OR-WA (30%).

In unaffordable areas, households can adjust their homebuying decisions in two ways, either by curtailing non-housing expenditures or foregoing the home purchase altogether. The latter explains why homebuying demand has been slowing in areas with unaffordable markets.

To sum up, homebuying demand has slowed nationally, but in states with strong job growth and affordable house prices, REALTORS® reported that homebuying demand remains “strong” compared to one year ago. The Federal Operations Market Committee is calibrating the increase in interest rates to rein in inflationary pressures without undermining economic and job growth.


[1] In a monthly survey of REALTORS®, respondents are asked “Compared to the same month last year, how would you rate the past month’s traffic in neighborhood(s) or area(s) where you make most of your sales?” Respondents rate buyer traffic as “Stronger” (100), “Stable” (50), or “Weaker” (0) and the responses are compiled into a diffusion index. An index greater (lower) than 50 means more (fewer) respondents reported “stronger” than “weaker” conditions in the reference month compared to the conditions in the same month last year. A higher value of the index in any reference month compared to the value in another reference month means a larger fraction of respondents reported “stronger” conditions in the former period than the fraction of respondents who reported “stronger” in the latter period. At the state-level, the index is calculated based on three months of survey data to increase the sample size, although small states such as Alaska, North Dakota, Maine, Vermont, and the District of Columbia may still have less than 30 responses.  

[2] I estimated the monthly mortgage payment on the NAR median price as of 2018 Q2, a 10 percent down payment, and the 2018 Q2 30-year fixed mortgage rate of 4.54 percent based on the Freddie Mac survey. I estimate the median household income by inflating the 2017 median household income based on the 2017 American Community Survey by the ratio of the average weekly wage for 2018 Q2 to the average weekly wage for 2017 Q2.

[3] HUD defines a household as cost-burdened if the households spends more than 30 percent of income on housing expenditures that includes mortgage payment, mortgage and home insurance, home maintenance, and utilities. However, because I am only calculating principal and interest, I am using the conservative benchmark of 25 percent, following NAR’s benchmark for estimating the Housing Affordability Index (HAI).

[4] Based on Haver Analytics platform calculations. Haver Analytics calculates the difference as the change in the quarter’s average of total non-farm employment in 2018 Q2 to the quarter’s average of total non-farm employment in 2017 Q2 using the DIFF method for 4 quarters (one year).

Recent Posts