By Philip Garboden
County breakdown update: https://uhero.hawaii.edu/county-results-from-small-business-survey/
Right now, everyone is making sacrifices. But as with any natural disaster, some people and places are being hit harder than others.
To better understand these dynamics, UHERO partnered with the Chamber of Commerce Hawaii and a dozen other industry associations across the state to field a survey of the impact of the COVID-19 shutdown on businesses and their employees [1]. In just a week, we collected data from 623 businesses. Thanks to each of you for taking the time to complete the survey! All business responses are confidential.
These data provide otherwise unavailable insight into the impacts of the economic crisis. They are already being used to inform our forecasting work and contain many interesting findings. In this initial post, we focus on the number one question: who is most affected by the shutdown?
Industry losses are large, and concentrated where you would expect them
The survey collected information on the number of employees each company employed in January of this year and the number that were still on payroll in April. We also asked businesses to provide their revenue in 2019 and to estimate what it will look like in 2020 – no easy task given the ever-changing nature of the virus response.
In total, businesses reported reducing their workforces by about 220,000 full- and part-time workers during the COVID-19 economic shutdown. A shocking figure, but not surprising given the number of unemployment claims being processed by the State.
Table 1: Employment and Revenue Impacts By Sector
| Employees | Revenue | ||||||
| Percent of Revenue from Tourism | January, 2020 | Change Jan to April | % Change | 2019 (Mean) | Anticipated Change 2020 | % Change | |
| Accommodation | 90.88% | 8,268 | -6,369 | -77.03% | $2,139,409 | -$1,049,091 | -49.04% |
| Administrative and support and waste management | — | — | — | — | — | — | |
| Agriculture, forestry, fishing and hunting* | 4.92% | 4,745 | -9,91 | -20.89% | $2,385,298 | -$401,047 | -16.81% |
| Arts, entertainment, and recreation | 89.80% | 8,823 | -7,269 | -82.39% | $2,015,843 | -$1,306,040 | -64.79% |
| Construction | 9.74% | 34,366 | -9,326 | -27.14% | $6,349,500 | -$178,731 | -2.81% |
| Educational services | 0.83% | 13,240 | -7,129 | -53.85% | $2,031,167 | -$825,000 | -40.62% |
| Finance and insurance | 23.41% | 15,950 | -102 | -0.64% | $8,574,861 | -$1,199,361 | -13.99% |
| Food services | 38.93% | 77,187 | -42,041 | -54.47% | $4,981,919 | -$2,134,973 | -42.85% |
| Industries not classified | 58.91% | 1,285 | -550 | -42.75% | $1,551,752 | -$693,836 | -44.71% |
| Information | — | — | — | — | — | — | |
| Management of companies and enterprises | — | — | — | — | — | — | |
| Manufacturing | 36.19% | 13,073 | -6,081 | -46.52% | $17,750,992 | -$5,432,462 | -30.60% |
| Other services (except public administration) | 43.48% | 12,092 | -7,370 | -60.95% | $2,998,851 | -$1,215,716 | -40.54% |
| Professional, scientific, and technical services | 10.69% | 19,826 | -1,897 | -9.57% | $1,447,861 | -$281,891 | -19.47% |
| Real estate and rental and leasing | 23.70% | 10,903 | -2,044 | -18.75% | $869,842 | -$214,763 | -24.69% |
| Retail trade | 66.99% | 58,760 | -45,012 | -76.60% | $3,600,375 | -$1,502,469 | -41.73% |
| Transportation and warehousing | 20.48% | 28,562 | -1,654 | -5.79% | $7,599,416 | -$2,409,262 | -31.70% |
| Utilities | — | — | — | — | — | — | |
| Wholesale trade | 60.31% | 16,364 | -4,576 | -27.96% | $4,866,750 | -$1,456,125 | -29.92% |
| Note: Sectors with fewer than five responses have been removed. *The agriculture number originally reported included an outlier observation which, upon examination, we do not consider reasonable. This line reflects the removal of that observation. | |||||||
As one would expect, these hardships are not evenly spread across industries. The hardest hit in terms of full-time employee reductions were accommodations (hotels), down 83%, and retail businesses, down 76%. Food, educational, and other services all also saw reductions north of 50% of their pre-crisis full-time workforces.
The anticipated changes in 2020 revenue, while at best guesstimates at this time, provide a sense of employer optimism or pessimism. They tell a similar story to the job loss reports, with a couple of important exceptions. Owners of retail businesses, one of the largest losers in terms of employees, were more optimistic about overall revenue, estimating just a 19% drop. In contrast, management companies are not experiencing large reductions in staff so far, but expect a substantial decline in revenue for the year overall.
In total, about a third (31%) of businesses reported that their revenue has been reduced to essentially zero during the shutdown, including 56% of hotels, 46% of restaurants, and 44% of retail.
The most vulnerable workers are also the most impacted
Within businesses, there are also differences in which employees have been let go. Unfortunately, the trend here suggests that many of the employees who have lost wages during this period are some of the most vulnerable – what are sometimes referred to as “ALICE” families (Asset Limited, Income Constrained, Employed), who have been shown in recent studies to be the most affected by Hawaii’s cost of living even under normal circumstances.
Table 2: Percentage of Jobs Lost by Job Type and Salary Range
| % Change in Total Jobs (Jan to Apr 2020) | |
| Full-time | 41.69% |
| Part-time | 51.43% |
| Full-time by Salary Range | |
| <$30k Salary | 30.82% |
| $30-50k | 35.04% |
| $50-100k | 30.50% |
| $100-200k | 28.38% |
| >$200k | 22.41% |
| Note: While rare, some changes may reflect movement between categories rather than job eliminations. | |
Part-time jobs were more likely to have been lost than full-time (56% compared to 43%). Among full time employees, those earning less than $50,000 per year were more likely to lose their position (35%) compared to 30% of those earning $50-$100k, 26% of those earning $100-$200k, and 16% of those making over $200k.
Oahu has the numbers, but tourism is clobbering the Neighbor Islands
Of course the largest absolute number of job losses has been on Oahu, but Oahu job losses as a percentage of their January level (38%) were significantly lower than for businesses located on the Big Island (48%), Kauai (52%), and Maui (58%). This is consistent with the larger share of tourism-reliant companies on the Neighbor Islands.
What do we know so far?
While this is a first look at the numbers, these statistics suggest several things:
First, businesses are suffering mightily, having already made painful cuts in staff, and they anticipate a very rough 2020 financially.
The good news is that most of the businesses surveyed anticipate being able to open and staff-up as soon as it is safe to do so. Indeed 60% said they could return to full staff almost immediately, with the rest phasing in as tourism returns to the Islands.
Unfortunately, the data we collected suggest that the Small Business Administration PPP forgivable loan program may not be enough to sustain many businesses. It will certainly help increase the retention of workers, primarily in part-time positions. But 24% of the businesses surveyed said that if no additional support is made available to them, they do not see their business surviving, and another 32 percent said they anticipate needing to make more staff cuts in order to survive.
Finally, the impact of the crisis will hit hardest the very people and places who are already the most vulnerable. While we did not find statistically meaningful differences in outlook based on gender or ethnicity of the business owner, we do show that the employees who have lost their jobs are likely those already struggling to make it. This will have wide-ranging consequences for these families and the need for supportive social services.
Update: County Results from Small Business Survey
A note on the survey analysis process: As noted above, we received hundreds of responses in quite a short amount of time. It would be ideal if these responses were entirely random, but the reality of any online data collection is that some types of business owners were more likely to hear about the survey and fill it out. To correct for differences in industry response rates, we have reweighted responses so that the total number of employees reported by businesses in the survey matches the totals for each business sector as reported by the Bureau of Labor Statistics. Overall, this weighting was relatively successful: the number of layoffs estimated from the weighted data is roughly in line with the number of unemployment claims filed over the past several weeks.
In interpreting percentage changes, note that our survey did not collect data on public sector employees or any of the large medical systems. Thus, it should not be generalized to all jobs in the state. The goal of the survey is to produce rough estimates of policy relevant statistics quickly, not to replace or challenge any official data from administrative records.
BLOG POSTS ARE PRELIMINARY MATERIALS CIRCULATED TO STIMULATE DISCUSSION AND CRITICAL COMMENT. THE VIEWS EXPRESSED ARE THOSE OF THE INDIVIDUAL AUTHORS. WHILE BLOG POSTS BENEFIT FROM ACTIVE UHERO DISCUSSION, THEY HAVE NOT UNDERGONE FORMAL ACADEMIC PEER REVIEW.
[1] Thanks to Chamber of Commerce Hawaii, The Hawaii Island, Kauai and Maui Chambers of Commerce, The Retail Merchants Association of Hawaii, the Pacific Resource Partnership, Hawaii State Bar Association, Hawaii Restaurant Association, the Chinese Chamber, Kalihi Business Association, the Hawaii Foreign-Trade Zone, and the Hawaii Food Manufacturing Association for help promoting the survey.
Photo by Peter Fuleky.