PPP loans approved by day
The loans from the Paycheck Protection Program (PPP) aim to address the economic consequences of the pandemic with payroll support for small businesses. The first run of the program distributed two rounds of funding from April to August; the latest stimulus package allocates funding to restart it. Earlier in December, the Small Business Administration (SBA) released detailed information on loan approvals through August 8, when the program initially ended.
Importantly, the loan-level dataset does not account for all the approvals reported by the SBA in its summary statement of the program, missing about 55,300 loans representing almost $2.1 billion. There is also incomplete reporting among the loans included in the data, with around 1% missing at least state, city, or zip code information. But this dataset, when used along with the SBA summary reporting, still provides the most complete accounting to date of the program.
Compared to first-round data, second-round PPP distributions evened out the share of aid across different states. Washington, DC and North Dakota remain the places with the most loan dollars per capita overall. However, Florida and California — two states with fewer approved loan dollars per capita in round one — received more in second-round PPP loans. These numbers come from the SBA summary report.
Returning to the loan-level dataset, counties in middle America — from North Dakota down to Texas — received the most PPP loans relative to population size. Among other factors, this might reflect a higher ratio of small businesses to people in this section of the country. These same counties have historically reported high rates of small business jobs as a share of total employment.
The loan-level data also shows how PPP loans varied across American cities. Out of the nation’s 30 most populous cities, San Francisco received the most money compared to its size, with an average of $3,900 dollars per person. In comparison, Detroit averaged around $970 per person, and El Paso, TX averaged just under $920.
The data does not give a definitive answer. But according to overview reports from the SBA as well as state-level employment data from the Bureau of Labor Statistics (BLS), more PPP dollars did go to states that faced more economic strain between January and August. States with high job loss relative to their population, like Hawaii and New York, as well as Washington, DC, received more PPP aid per 100 people than did states that reported fewer job losses, such as Mississippi and Idaho.
The distribution of PPP dollars was not consistent with the level of coronavirus spread within the 30 most populous counties. Take New York, where the Bronx reported the highest cumulative COVID-19 case count through August 8 and the least in PPP loans, at 3,500 cases and around $77 million per 100,000 people. In contrast, Manhattan averaged 1,900 cases and over $669 million per 100,000. Queens, Brooklyn, and the two counties of Long Island fell somewhere in the middle.
Among other factors, this might reflect how the health and economic consequences of the pandemic, while connected, can affect different places. With reduced travel, more people staying at home, and fears about where the pandemic might spread next, even places with few COVID-19 cases can face damage to local economies. And as mentioned before, population density and small business density also vary across counties.
Most loan approvals also happened early in the pandemic, with around 56% of the loans and 81% of the dollars approved in April. New York was the primary coronavirus hot spot in the spring — and it accounts for three of the four counties with the most PPP dollars per capita. While Miami-Dade County recorded the most cumulative COVID-19 cases per capita as of August 8, its cases only began to grow later in the summer. It received the 10th most dollars per capita among the 30 largest counties.
Of the nation’s 30 most populous cities, Detroit and El Paso received the fewest dollars relative to their population. Both count white and non-Hispanic Americans as fewer than 15% of their residents.
As the pandemic continues, with higher mortality rates for Black and Hispanic Americans, such statistics raise the potential of additional disparities when it comes to the economic fallout.
Get more facts on PPP and access to the data used in this article. Datasets are linked under "What is this data?" in the "Questions to ask" section.
The loan-level data does not explicitly identify counties, instead using zip codes and city names. Because zip codes and counties are not perfect matches, a loan from a given zip code can sometimes belong to one of several counties. To accommodate for this, loans from zip codes that cross multiple county boundaries were divided equally between those counties. Zip codes with this issue were in the minority for the top 30 counties analyzed, with just 126 zip codes affected out of a total of 3,734.
Keep up with the latest data and most popular content.