CROWE Data Briefs are short, regular reports meant to highlight timely and important topics. The format is short, punchy, and centered around a few key pieces of data. Data briefs bring economic and policy discussions down to the personal level, showing how economic issues affect individuals and communities.
This is an accordion element with a series of buttons that open and close related content panels.
"Wisconsin’s labor market and COVID-19" by Kim Ruhl, July 24, 2020 (updated Oct 2, 2020)
This brief reports labor-market indicators for Wisconsin to demonstrate the effects of COVID-19 on labor supply and demand. This report is part of a larger effort at CROWE to document and analyze the economic fallout of the COVID-19 pandemic.
"The COVID-19 Recovery Has Stalled" by Noah Williams, July 24, 2020
After a rapid and deep crash with the onset of the COVID-19 pandemic and associated lockdowns, economic activity rebounded sharply from mid-April through the end of June across the country. However activity has fallen from its peaks in early July and has been flat over the past ten days or so, roughly in line with the spread of the virus and the reimposition of more stringent public health restrictions in different locations. In many states economic activity by mid-July was back where it was roughly one month earlier, if not lower.
"The Economic Impact of the Wisconsin Supreme Court Ruling Invalidating the State’s Safer at Home Order" by Junjie Guo, July 3, 2020
In a 4-3 ruling, the Wisconsin Supreme Court invalidated the state’s Safer at Home order on May 13, 2020, allowing most non-essential businesses to open immediately in most locations. Using several real-time measures, I find broad evidence suggesting a positive impact of the Supreme Court decision on economic activity. From May 13 to May 27, relative to states where non-essential businesses were shut down, Wisconsin experienced a larger increase in the number of small businesses open (7.8 ppts), net revenue for small businesses (4.6 ppts), employment of low-income workers (0.8 ppts), earnings of lowincome workers (1.6 ppts), individual mobility as measured by GPS data on time spent outside residential locations (3.2 ppts) and consumer credit/debit card spending (3.1 ppts). The impact appears to be larger for sectors closely related to accommodation and food services. However, because these sectors were hit especially hard initially, the gap in year-over-year activity for these sectors was still larger than the gap for the overall economy two weeks after the Supreme Court decision.
Qualitatively, the evidence suggests the state’s Safer at Home order was binding, and its invalidation contributed to economic recovery. Quantitatively, the impact is not very large, reducing the initial gap in year-over-year activity on May 13 by less than a quarter for majority of the economic measures, and the responses from households seem to be slower and smaller than firms. Overall, the evidence suggests a modest role of the Safer at Home order on economic activity, consistent with my previous analysis of the initial impact of the order’s implementation as well as similar analyses by others.
"Protest Dynamics: Evidence from Foot Traffic data" by Noah Williams, June 26, 2020
Following the killing of George Floyd in Minneapolis on May 25, there have been substantial protests in many cities around the country, and indeed around the world. In this brief I use foot traffic data to analyze the dynamics of the protests at different locations around the United States. In both Minneapolis and Washington, DC, I find protests growing in scale to a peak, and diminishing thereafter. Since early June, overall activity has dropped substantially in the zip code in Minneapolis that was at the heart of the protests, likely due to sustained damage at area businesses. In Washington, protest activity around the White House started later, growing to a peak on June 6. Activity has declined since then, but remains elevated relative to April and May, and has spiked on weekends. Overall, I find that the protests led to isolated spikes in activity at particular locations. But the protests did not substantially impact overall measures of activity in the metro areas where they took place.
"The Wisconsin Economy During COVID-19: Lockdown and Reopening" by Noah Williams, June 11, 2020 (updated Oct 2, 2020)
This brief summarizes data on the Wisconsin economy since the onset of the COVID-19 pandemic. In particular, I analyze economic activity using foot traffic at commercial locations around the state.
Since the last report on September 11 (reflecting data through September 6), the Wisconsin activity has experienced a slowing of economic activity. After a long period of plateau, stretching from early July through early September, different measures of activity all showed a sharp decline in September. Although there are signs that activity has rebounded slightly over the past week, activity remains below the previous few months, reverting to levels last seen in June. Over the last three weeks, overall foot traffic from SafeGraph data has fallen by 8 percentage points, with a decline of 11 points in the hardest-hit accommodations and food sector. Similarly, small business employment from Homebase data has fallen by about 7 percentage points since the end of August, and about 10 percentage points in the food and drink sector.
The slowing in activity comes as COVID-19 activity in Wisconsin spiked to new highs, with record numbers of cases and broad spread in the virus across the state. In addition, the timing of the year-over-year slump in September coincides with the back-to-school period and so reflects diminished activity from purchases as well as fewer college students returning to campuses. This latter effect is especially notable in Madison, where foot traffic fell to roughly 50% of 2019 levels in mid-September before recovering slightly of 55% in the last week of the sample.
"Business Formation during the COVID-19 Pandemic" by Simeon Alder, June 11, 2020 (updated Sep 25, 2020)
The COVID-19 pandemic has led to widespread disruptions in the U.S. economy, especially in labor markets. In addition to the widely reported rise in unemployment and job losses, there is timely and high-frequency data that shows a significant disruption in early-stage business formation. Using the Census Bureau’s Business Formation Statistics (BFS), which tracks applications for Employer Identification Numbers (EINs) at the weekly frequency, this report explores the link between labor market fluctuations and early stage business formation. We find that changes in employment and labor force participation rates play a fairly limited role in accounting for business formation in “normal” times. Both the employment and labor force participation rates change very slowly and virtually all growth in applications per capita is accounted for by a rise in applications per worker.
During the COVID-19 pandemic, however, labor market disruptions of a truly unprecedented magnitude play a far more prominent role in explaining the collapse of business formation in the US. Typically, changes in the employment and labor force participation rate account for half or more of the observed decline in applications per capita. While there are signs of a recovery in the number of business applications in the second half of May, it is too early to connect these developments to state-level labor market data.
One of our concerns is that a prolonged decline in the labor force participation rate may have a scarring effect on business formation. While this effect is not specific to Wisconsin, it would exacerbate the state’s previously documented lack of business dynamism and the Center is therefore monitoring future labor market and business formation developments closely.
The third update to the “Business Formation during the COVID-19 Pandemic” report updates the business formation statistics to include calendar weeks 29 through 33 and the state-level labor market data for the month of July. Labor market conditions and business applications are linked using the same accounting approach as the original report and the two previous updates. The rebound in business formation continued through the first two weeks of August and year-to-date totals now exceed the corresponding cumulative numbers from previous years, both at the national level and in virtually all Midwestern states. Local and national labor market conditions, on the other hand, continue to be poor by historical standards. Modest year-on-year gains in the employment rate (and hence a corresponding drop in the unemployment rate) since June are partially offset by a deterioration in the labor force participation rate.
For a more detailed discussion of the underlying methodology and data sources, the interested reader may want to review the original report, released on June 12, 2020.
"Reopening the Economy: Early Evidence from Georgia" by Noah Williams, May 14, 2020 (updated May 29, 2020)
After implementing stay-at-home orders to slow the spread of the COVID-19, over the past couple of weeks, a number of states have moved toward “reopening” their economies. Georgia has had one of the earliest and most comprehensive reopening plans. I compare economic activity in Georgia and Wisconsin, which had remained under a stay-at home order until a recent state Supreme Court decision. I find that activity in both states was very similar during the early pandemic and stay-at-home orders. However the removal of restrictions in Georgia has been followed by a 10 percentage point increase in activity over the past three weeks, with slightly larger gains in industries particularly targeted for earlier reopening. Activity in Georgia remains 35% below 2019 levels, thus the gains after reopening accounted for roughly 22% of the gap (10 points out of 45), consistent with previous causal estimates of the impact of stay-at-home orders.
"Measuring proximity to others in the workplace" by Kim Ruhl, May 5, 2020
As governments grapple with the costs of social distancing, a discussion has emerged about how “reopening” the economy might work. Most proposals suggest a staged reopening, beginning with relatively low-risk sectors. How can we identify low-risk sectors? In this brief we use data on the proximity to others while working to study this issue.
"The Impact of Statewide Stay-at-Home Orders: Estimating the Heterogeneous Effects Using GPS Data from Mobile Devices" by Junjie Guo, April 29, 2020
In response to the spread of COVID-19, most states in the U.S. have ordered nonessential businesses to close and residents to not leave home unless necessary. This paper estimates the impact of these stay-at-home orders on the mobility of Americans using GPS data from mobile devices. Recognizing the heterogeneity in both mobility and the orders across states, I use the synthetic control method at the state level to estimate the impact of individual orders instead of pooling them together for an average effect. I find the impact does vary significantly across orders/states. For example, the estimates suggest that the orders in Michigan and Wisconsin increased the fraction of their residents at home all day by about 5.5 and 4 percentage points, respectively, while the corresponding estimate for Ohio is small and insignificant. In addition to the effectiveness of the orders in limiting the spread of COVID-19, these estimates are also informative of the responsiveness of Americans when the orders would be lifted.
"Consumer Responses to the COVID-19 Pandemic" by Noah Williams, April 16, 2020 (latest update : April 23, 2020)
As the COVID-19 pandemic has spread across the United States, consumers have changed their spending habits dramatically. Initially, the changes began with people canceling inessential travel, limiting attendance at large gatherings, generally limiting social exposure, and moving from purchases in-store to online. Social distancing guidelines in most of the US tightened throughout the month of March, with the issuance of public emergency declarations, and an increasing number of states implementing “stay-at-home” orders closing down non-essential businesses. These changes, and the increasing severity of the crisis, accelerated the changes in consumer behavior, and also led to spikes in certain purchases, particularly at grocery stores. By the end of March and into April, consumer behavior has roughly stabilized, but with vastly different buying patterns and behavior compared to only a few weeks ago.
In this brief I analyze consumption trends nationwide and in Wisconsin, using a new data source of weekly transactions. I find that the recent monthly decline of roughly 9% in national retail sales masks a 10% increase mid-March and a 20% decline by the end of the month. There was a spike in grocery store sales of 80% nationwide and 100% in Wisconsin mid-March, driven as much by a growth in sales-per-transaction as transactions. The consumption decline overall was cushioned by increasing on-line sales, which also reallocated sales. In Wisconsin, total sales were down 15% at the end of March, but in-store sales were down 30%, with online sales up 20%.
"Air Traffic Data as a Proxy for Economic Activity in the Transportation Sector" by Simeon Alder, April 8, 2020
The imposition of numerous social distancing measures in response to the COVID-19 pandemic has curtailed economic activity significantly. The impact, however, varies across industries and one of the challenges at this time is to measure the impact in a timely manner. One sector that has been affected significantly is air transportation. Measures ranging from outright quarantines to milder “Safer at Home” directives have been associated with a sharp decline in mobility – both internationally and domestically – and the number of airline passengers has fallen sharply in recent weeks.
In this brief we use daily departure information at Wisconsin’s three major commercial airports, General Mitchell International Airport in Milwaukee, Dane County Regional Airport in Madison, and Green Bay Austin Strobel International Airport in Green Bay to gauge the decline in activity between the beginning of March and April 6, 2020. We compare these trends to broader national air traffic patterns and some international developments.
"The Economic Impact of COVID-19 on the Chinese Economy" by Chang Liu, April 6, 2020
The COVID-19 pandemic has now spread across the globe, causing significant economic disruption all over the world. China was the first and one of the hardest-hit countries in this global fight. The city of Wuhan, the first epicenter of this pandemic, with a population of 11 million, was completely shut down on Jan. 23, 2020, which WHO called “unprecedented in public health history”. Soon after that, cities in Hubei, of which Wuhan is the capital city, together with the rest of China, imposed similar lockdown restrictions, albeit less restrictive. Due to these quarantine regulations, by late February officially reported new cases of coronavirus had turned zero outside Hubei; in mid-March, total officially reported new domestic cases in mainland China turned to zero.
This fight, however, comes with substantial economic costs: businesses had to shut down, transportation scaled down, workers and consumers stuck at home and so on. With the spread of coronavirus under control, restrictions on businesses were gradually lifted throughout March.
In this report, I use the newly available data to investigate how the COVID-19 outbreak and the associated quarantine policies have affected the Chinese economy from various perspectives. Specifically, I look at the 8 groups of economic indicators covering production, investment, consumption/sales and imports and exports. Almost all of them indicate a plummet in economic activity where the year-over-year growth rate dropped from about 10% before the pandemic to -20% or lower in February 2020. Fortunately, the limited amount of data available for March indicates a rebound in economic activity, which suggests that the negative impact may be short-lived.
"Measuring Wisconsin Economic Activity Using Foot Traffic Data" by Noah Williams, latest update April 8, 2020
The COVID-19 pandemic is leading to unprecedented social and economic disruptions around the globe. The economies in many locations have ground to a halt, as social distancing measures to slow the spread of the virus have increasingly led to businesses being shut down and workers ordered to shelter in place. While there has been an immense reduction in economic activity, there have been relatively few measures of just how severe the impact has been.
In this brief I analyze a measure of economic activity using a new data source of foot-traffic in commercial locations. I focus on year-over-year same-location changes in the state of Wisconsin, and find that there has been roughly a 52% drop in overall activity during the last week of March 2020 compared to 2019. The declines have been even more severe in some industries, with 76% drops for hotels and 71% for restaurants. Grocery stores are the only retail sector seeing relatively strong activity, with activity off only 7%. While non-essential retail stores have closed, grocery stores have remained open and activity at certain times has spiked. With the closing of UW-Madison, the Madison metro area has seen an even larger 65% decline in activity. Activity has stabilized, at a low level, over the last two weeks.
"The effects of COVID-19 on Wisconsin’s workers and firms" by Kim Ruhl, latest update June 26, 2020
The “social distancing” required as part of the response to COVID-19 has created a sharp increase in Wisconsin workers seeking unemployment insurance. We focus on initial unemployment claims, which measure applications for benefits from workers who were not currently receiving benefits.
"Unemployment Benefits under the Federal COVID-19 Relief Package" by Noah Williams, March 27, 2020
As the COVID-19 pandemic has spread throughout the United States, it has led to mass shutdowns of businesses with many states being placed under shelter-in-place orders. This is leading to an unprecedented increase in unemployment, with initial unemployment claims hitting 3.3 million on the week of 3/21/20, which will likely increase in the coming weeks. To provide relief for these unemployed workers, the federal government is implementing an aid package which supplements state unemployment insurance benefits with an additional $600 per week in “Federal Pandemic Unemployment Compensation” through 7/31/2020.
This brief compares the unemployment benefits compensation through existing state systems and the new federal compensation to average weekly wages in each state. I find that maximum unemployment benefits would exceed 90% of average weekly wages in all states, with the ratio substantially higher in most states.
"The Geographic Distribution of Workers Most At Risk Economically When Ordered To Stay At Home" by Junjie Guo, March 26, 2020
On March 24, 2020, at the direction of Governor Tony Evers, Wisconsin issued a “safer at home” order requiring residents not to leave their home unless necessary. Similar orders to lock down the economy in an attempt to limit the spread of COVID-19 have been issued elsewhere. Different from many other countries like China, Italy and most recently the U.K., the federal government in the U.S. hasn’t issued such an order nationally, and President Trump recently expressed his willingness to move in the opposite direction and “have the country opened up … by Easter” after his direction of 15-day social distancing ends next week, a timeline that is dramatically sooner than what many public-health experts have recommended. Instead, some state and local governments have been issuing lockdown orders on their own. At the time of this writing (11:50 pm CST on 3/25/2020), 196 million Americans in 21 states, 37 counties and 16 cities are being urged to stay at home, according to a real-time tracker by the New York Times, which also noted that “A few states – Kentucky, Maryland and Nevada, for example – have walked up to the line, closing down all non-essential businesses but not issuing formal orders for people to stay home”.
"Forecasting Initial Unemployment Claims using Google Searches" by Noah Williams, March 23, 2020
Weekly Google search trends related to unemployment provide an accurate forecast of initial unemployment claims. For the current week (ending March 21), I forecast that US initial unemployment claims will top 1.6 million, which is roughly twice the previous record high.
"COVID-19, Industry Mix and the Growth of Initial UI Claims across States" by Junjie Guo, March 23, 2020
In its latest news release on unemployment insurance (UI) weekly claims, the U.S. Department of Labor (DOL) reported that, relative to the prior week, the number of initial claims increased significantly by 70,000 in the week ending on March 14, 2020. The report noted that the increase “are clearly attributable to impacts from the COVID-19 virus. A number of states specifically cited COVID-19 related layoffs, while many states reported increased layoffs in service related industries broadly and in the accommodation and food services industries specifically, as well as in the transportation and warehousing industry, whether COVID-19 was identified directly or not.”
We evaluate this assessment quantitatively.
"The Uneven Effects of Chinese Tariffs" by Kim Ruhl, November 12, 2019
As part of the escalating trade war, the Chinese government increased substantially the tariffs it levies on a range of products exported from the United States. Trade policy is applied at a national level—a good entering China from the United States is tariffed the same regardless of where it was produced within the United States. The United States, however, is geographically diverse and the effects of the Chinese tariffs fall unevenly on different parts of the country. In early 2018, the United States levied tariffs against solar panels, washing machines, steel and Aluminum—beginning the ongoing trade war.
"The Dodd-Frank Act and Small Bank Creation", by Kim Ruhl, May 15, 2019
In the aftermath of the financial crisis, Congress passed the Dodd-Frank Act which increased capital requirements and regulation of banks. Following the Act, there has been a reduction in the creation of small banks, who have faced a large burden under the Act. Small banking is most important in poorer, rural areas, which suggests that these areas would be most affected by changes in the health of the small-bank sector.