U.S. Retail Sales Record Plunge, Weighed Down by Coronavirus Shutdowns

• Consumers stay home and do without most new products.  

• Apparel & furniture sales are notably hard hit.

Total retail sales including food service establishments plunged 8.7% (-6.2% y/y) during March after weakening 0.4% in February, revised from 0.5%. A 5.5% decline had been expected in the Action Economics Forecast Survey. Retail sales excluding motor vehicles & parts fell 4.5% (-1.7% y/y) following an unrevised 0.4% easing. A 2.1% decline had been anticipated. Motor vehicle & parts dealerships posted a 25.6% sales decline (-23.7% y/y) as unit sales of motor vehicles weakened by one-third m/m and y/y. 

Retail sales fell 6.2% last month (-3.8% y/y) after slipping 0.5% in February. Apparel & accessory store sales plummeted 50.5% (-50.7% y/y), down sharply for the third straight month. Sales at furniture & home furnishings stores weakened 26.8% (-24.6% y/y) after easing 0.9%. Sporting goods, hobby, book & music store sales fell by roughly one-quarter both m/m and y/y. Electronics & appliance retailers realized a 15.1% decline in sales (-15.9% y/y) which came after a 0.9% drop. Restaurant and bar sales declined 26.5% (-23.0% y/y) with widespread closures of these establishments. Gasoline & service station sales fell 17.2% (-18.0% y/y) as prices dropped and driving was limited.

Partially offsetting these declines was a 25.6% surge (28.0% y/y) in food & beverage store sales as eating home became a necessity. General merchandise stores realized a 6.4% (7.5% y/y) surge in sales. Sales of nonstore retailers also strengthened 3.1% (9.7% y/y), the largest monthly increase since January of 2019. The gain followed two months of modest increase. Building materials & garden store sales improved 1.3% (7.6% y/y).

In another nondiscretionary sales category, health & personal care product stores improved 4.3% both m/m and y/y, following a 0.4% decline.

The retail sales data can be found in Haver's USECON database. The Action Economics forecast is in the AS1REPNA database.

Mitigating COVID-19 Effects with Conventional Monetary Policy from the Federal Reserve Bank of San Francisco is available here https://www.frbsf.org/economic-research/files/el2020-09.pdf

The Mortgage Bankers Association reported that its Mortgage Loan Application Index declined 17.9% (+51.0% y/y) in the week ended April 4 and reversed the prior week's 15.3% gain. Purchase applications dropped 12.2% w/w and by one-third y/y. Applications for refinancing fell 19.4% w/w (+143.5% y/y) after having risen by one-quarter in the previous week.

The effective interest rate on a 15-year fixed-rate mortgage slipped to 3.10% from 3.12% in the prior period. The effective interest rate on the 30-year fixed-rate mortgage held steady at 3.57%. The effective rate on a 30-year Jumbo mortgage rose to 3.95% from 3.93%. The rate on the 5-year adjustable rate mortgage increased to 3.38% from 3.34%.

The average mortgage loan size fell sharply to $313,200 from $323,700 in the previous week. The average loan size of refinancings declined to $312,400 from $322,800. The average loan size for purchases dropped to $315,300 from $326,500, the lowest price level since the second week of February 2019.

Applications for fixed-rate loans increased 58.1% y/y, while applications for adjustable rate loans fell 34.6% y/y.

The survey covers over 75% of all U.S. retail residential mortgage applications and has been conducted weekly since 1990. Respondents include mortgage bankers, commercial banks and thrifts. The base period and value for each index is March 16, 1990=100. The figures for weekly mortgage applications and interest rates are available in Haver's SURVEYW database.

The Composite Housing Market Index from the National Association of Home Builders-Wells Fargo fell 2.7% m/m to 72 in March from 74 in February for the third consecutive monthly decline. Still, the index is up 16.1% from a year ago and continues to fluctuate near its highest level in two decades. A reading of 73 had been expected in the INFORMA Global Markets survey. The NAHB figures are seasonally adjusted. Over the past 15 years, there has been a 70% correlation between the y/y change in the home builders index and the y/y change in new plus existing home sales.

The index of present sales conditions declined 2.5% m/m to 79 (+16.2% y/y). The index of expected conditions in the next six months fell 5.1% m/m to 75 (+4.2% y/y), its lowest reading since September. The index measuring traffic of prospective buyers slipped 1.8% m/m to 56 (+27.3% y/y), its second consecutive modest decline from January’s expansion high of 58.

It should be noted that half of the builder responses in the March index were collected prior to March 4, so the recent stock market declines and the rising economic impact of the coronavirus will likely be reflected more in next month’s report. Overall, 21% of builders in the survey reported some disruption in supply due to virus concerns in other countries, notably China. However, the incidence is higher (33%) among builders who responded to the survey after March 6, indicating that this is still an emerging issue.

Regional readings were generally lower in March. The index for the Northeast fell 3.0% m/m to 64. The index for the South declined 3.8% m/m to 76. And the index for the West decreased 3.7% m/m to 79. In contrast, the index for the Midwest increased a solid 6.3% m/m to 67, offsetting a similar decline in February.

The NAHB has compiled the Housing Market Index since 1985. It reflects survey questions which ask builders to rate sales and sales expectations as "good," "fair" or "poor" and traffic as "very high," "average" or "very low." The figures are diffusion indexes with values over 50 indicating a predominance of "good"/"very high" readings. In constructing the composite index, the weights assigned to the individual index components are: 0.5920 for single-family detached sales, present time, 0.1358 for single-family detached sales, next six months, and 0.2722 for traffic of prospective buyers. These data are included in Haver's SURVEYS database.