The Big Idea
Check, please!
Stephen Stanley | February 25, 2022
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
As a bellwether for the impact of the pandemic on consumer behavior, restaurants have worked pretty well. Each time the virus has surged, households have pulled back on activities that involve public contact, including travel and eating out. With Omicron fading in the rear-view mirror, the high-frequency data point to a substantial pickup in people eating at restaurants. It’s a signal that services spending more generally is likely on the cusp of a solid bounce after moderating in December and January.
Restaurant receipts
The Census Bureau’s monthly survey of retail sales covers a variety of types of retailers, all of whom sell goods. There are general merchandise stores, such as Wal-Mart and Target, clothing stores, home improvement outlets, gas stations, grocery stores, drug stores, and even nonstore retailers that mainly sell over the Internet. The only exception to the goods focus in the retail sales report is that the Census Bureau also collects data from restaurants.
Restaurants receipts slid late in 2020, when Covid heated up, and then recovered sharply in the first half of 2021 as the pandemic eased, two rounds of rebate checks filled household coffers and labor income surged. However, in the second half of the year, restaurant receipts leveled off in the fall. Omicron led to modest declines in restaurant receipts of 0.6% in December and 0.9% in January, which left the level of receipts flat on balance from July to January.
OpenTable data
Since the onset of the pandemic, economists have relied more than usual on high-frequency data, as economic activity has tended to move up and down in larger magnitudes and more quickly than in a normal economic environment. For the restaurant industry, the data that has drawn the most focus has been numbers from OpenTable measuring the quantity of seated diners in restaurants around the country. These figures are incredibly timely—daily with a single day lag—and are quite detailed, including national data for the US and several other countries as well as state-by-state tallies.
OpenTable reports their data as daily percent changes compared to the corresponding day in 2019 to escape the distortions caused by the pandemic in 2020 and 2021. Figures for 2020 are year-over-year but 2021 observations are compared with two years ago, and the readings for early 2022 are gauged against the corresponding day three years ago. Of course, daily data are subject to massive swings, as weather, floating holidays, and other factors can sharply affect the year-over-year comparisons for any given day. A 14-day moving average smooths most of those fluctuations (Exhibit 1).
Exhibit 1: OpenTable seated diners 2020-22 compared to. 2019
(Percent change 14-day moving average)
Source: OpenTable.
The OpenTable numbers show that after a sharp recovery in the spring and summer of 2020 as lockdowns eased, the winter 2020-21 Covid surge—and perhaps weather less conducive to outdoor dining—led to a sharp pullback in restaurant diners, taking the year-over-year decline from around 40% all the way down to almost 70%. In contrast, the dip seen in August 2021 related to the Delta wave was just a few percentage points.
As the chart shows, the deterioration caused by Omicron was much more extensive than during the Delta wave, as the OpenTable gauge softened from only slight declines just before November to drops of between 25% and 30% at the lows in mid-January, a weakening that corresponds to the back-to-back monthly decreases in restaurant receipts in the December and January retail sales reports.
However, restaurant bookings have bounced back as rapidly as the Omicron surge has dissipated. The trailing 2-week average returned to the pandemic-best of -1% by February 23 and increased into positive territory for the first time since COVID initially mushroomed in early 2020 on Friday. This suggests that restaurant receipts are likely to strengthen noticeably in February.
Monthly averages of the 14-day moving averages also tell a story. This measure is perhaps trailing to a degree, but it offers a bigger-picture view of the progress of restaurant diners, closer to being analogous to the monthly retail sales numbers (Exhibit 2).
Exhibit 2: OpenTable monthly averages
Source: OpenTable.
After a period of rough stability through the summer and fall—in the end, the Delta wave made little difference in restaurant attendance—the slide in diners in December and January was extensive. However, the bounceback has come quickly. As noted above, this gauge will tend to lag somewhat—it is a monthly average of a 14-day trailing average—so the February reading of -13% likely understates the degree to which restaurant reservations have already rebounded. In fact, the average over just the past seven days has already moved decisively into positive territory.
The post-Omicron bounceback in restaurant spending and presumably more broadly in services spending may be split between February and March. In any case, the main takeaway is that the consumer hunkered down for a time when Omicron raged but re-emerged quickly as soon as the virus moderated.