The Time is Now for Nowcasting

A big challenge for making budget and policy decisions in the time of COVID is the lack of real-time information on what is happening around us.  Many government measures, especially those reported episodically and based on surveys of households and businesses, feature lag times offering reduced informational value for the rapidly changing circumstances we are experiencing today.  Moreover, these measures are often unable to provide the granularity needed to understand what is going on in specific sectors, geographic areas, or within subgroups of populations.

For this reason there is heightened interest around the country in “high-frequency” reported data to obtain a better understanding of what is happening to economies on a real-time basis.  The distinguishing feature of this data reporting is that is uses information from private companies and industry groups on consumer spending, employment, business revenues and a host of other areas, all anonymized to protect privacy concerns.  By monitoring transactions and economic activity in real time, researchers can provide a better understanding of what is happening (a.k.a. “nowcasting”) and how policy solutions might be tailored for better impact and effect.

To support the economic recovery from COVID, a collaborative effort of university researchers based out of Harvard called “Opportunity Insights” has developed a freely available interactive website to provide a real time picture of national and state economies.  This Economic Tracker[1] combines anonymized data from private company partners to generate new statistics and new insights on how COVID is affecting the economy, businesses and people.  As the developers note, “rather than waiting weeks to see where the economy is falling and playing catch-up, the tracker offers the capacity to spot economic problems as they emerge and develop targeted, evidence-based policy responses – providing a powerful new tool for economic policy.”

Behind the Numbers – A Higher Income, Consumption Driven Recession

Government statistics show that COVID has led to a sharp reduction in GDP and an explosion in unemployment.  Most of the reduction in GDP has come from a major contraction in consumer spending (rather than business investment, government purchases or exports.)  In a new research paper[2] Opportunity Insights examined what is going on behind these reported numbers with respect to consumer spending, business shocks, and employment trends.  The research also generated some insights and conclusions on how government policies have influenced these outcomes.

• Spending -- Opportunity Insights found changing behaviors of high-income, service-consuming households are the primary source of economic contraction to date.  As of mid-June more than half of the reduction in spending since January has come from households in the top income quartile.  Only 5% has come from the lowest income quartile.  High-income households reduced their spending by 17%; the lowest quartile reduced spending by only 4%.  Unsurprisingly, most of the reduction in service spending stems from in-person services.  Services demanding less interpersonal interaction experienced far smaller losses.  Reductions in spending were found to be the greatest in high-income, high-density areas with higher rates of infection.  Together, these findings led researchers to conclude consumer spending has fallen primarily due to health concerns and the inability to supply services without health risks rather than lost income.
• Business Shocks – Continuing the theme of a “higher-income, consumption driven” recession, researchers found the impacts of COVID on small businesses are highly diverse with the greatest impacts occurring in wealthier, higher-income areas.  Nationally, more than half of the aggregate loss in small business revenues has come from businesses located in the top quartile of affluent zip codes; only 8% from the bottom quartile.  That heterogeneity exists even within cities.  Small business revenues fell by more than 70% between March and late April in most affluent city zip codes as compared to 30% in least affluent city zip codes.
• Employment – Researchers found that as businesses lost revenues, the shock was passed along to employees.  Once again workers in higher-income areas experiencing disproportionately more of the fallout.  In the most affluent zip codes, more than 65% of small business workers lost their jobs within two weeks of the beginning of the COVID crisis.  In contrast, in the least affluent zip codes, fewer than 30% lost their jobs.  Job postings also fell much more sharply in affluent areas.  As a result of this one-two combo of unemployment and reduced job prospects, low-income individuals working in these more affluent areas cut their own spending much more than low-income individuals working in less affluent areas. 

Much of this may change as the relationship between COVID and the economy evolves and safeguards such as the federal stimulus and “bonus unemployment” – which has protected a lot of lower income households – expires.  But as the authors conclude: “the initial impact of COVID 19 on economic activity appears to be largely driven by a reduction in service spending by higher income households due to health concerns which in turn affected businesses that cater to the rich – e.g. small businesses in affluent areas -- and ultimately reduced the incomes and expenditures of low-wage employees of those businesses.”

What Have Government Policies Accomplished

Real-time data also enables investigations into the impacts economic stabilization policies may or may not be having.  The researchers evaluated three such policies: state-ordered reopenings, federal stimulus payments to households, and the Paycheck Protection Program.

• Government Reopenings  -- For all the political sturm und drang surrounding state economy reopenings, the evidence suggests quite modest impacts on economic activity.  Spending and employment remained well below baseline levels even after reopenings and notably did not rise more rapidly in states that reopened earlier relative to comparable states that reopened later.  Moreover, spending and employment fell before these state level shutdowns were implemented.  Researchers concluded the key driver of reduction in spending is the virus itself rather than restrictions imposed by government, and reopenings only help if the public interprets them as a truly credible indicator of reduced health concerns.
• Stimulus Payments – Approximately $267 billion of direct payments have been made to 160 million people as part of the federal CARES Act.  Larger payments went to lower-income households, most of which were made on April 15.  Real-time data indicated the stimulus payments increased spending by low-income consumers immediately and substantially.  However, most of this spending went to the purchase of durable and non-durable goods rather than the service sectors hit hardest by the pandemic.   Thus, to date, researchers conclude the stimulus has offered little economic and employment benefits in communities and areas where job losses were largest.
• Paycheck Protection Program -- Loans targeting small businesses through the PPP appear to have had little effect on employment at small businesses to date.  The percentage changes in total earnings and hours worked are nearly identical for firms eligible and not eligible for the PPP.  Researchers offer three possible explanations: 1) businesses who took up loans did not intend to lay off their workers to begin with; 2) high take-up rates among firms providing professional scientific services despite low job losses in that service sector; and 3) loans flowed to areas with smaller employment losses.

A Look at How Minnesota Compares

Opportunity Insights’ Economic Tracker offers perspective on how well the state is navigating the crisis.  The accompanying table benchmarks the state of the Minnesota economy against the U.S. as a whole on several summary indicators since January. (As an aside we can’t help but note data included in the tracker from Google Mobility Reports finds the change in time spent away from home in Minnesota since January is now 0% meaning Minnesotans are in one sense “back to normal” compared to pre COVID days.  However, a closer look reveals continued reduced times at workplace and in transit is being completely offset by additional time spent in parks – a very Minnesotan finding.)

Source: Opportunity Insights Economic Tracker.  Reported data is from when article was being assembled (mid-June) and is not current at time of publication.

As the table shows, Minnesota is tracking with national declines in consumer spending albeit with some differences when the spending is disaggregated by income.  Small business revenue declines continue to be greater than the U.S. total.  Perhaps most disconcerting is that small business revenue declines in low income zip codes are well over twice the nation is experiencing (-30.8% vs. -13.6%).  From a sector perspective, Minnesota professional and business service small business revenue is slightly “outperforming” the nation’s decline while the state’s transportation and trade small business sector is experiencing dramatically more significant revenue losses than the national average (-25.4% vs. -4.4%).

Rebounds in the number of small business openings are lagging the U.S as a whole, and once again persistent closures in low income zip codes are primarily responsible for the difference.  Declines in Minnesota job postings are slightly larger than the nation in every sector with the exception of education and health services.  While job postings requiring “minimal education” are now actually up from the beginning of the year in the U.S., they are down over 10% in the state.

What is rather striking about these results is that the Minnesota experience appears to run counter to the nation as a whole in a significant way.  As previously noted, researchers from Opportunity Insights concluded small businesses in affluent areas have taken the brunt of the economic impact from COVID so far.  In Minnesota the opposite appears to be true – small businesses in the least affluent areas have by far taken the biggest hit.  In fact, since the Minnesota stay at home order ended, it appears every trackable economic metric on the website has shown some improvement except one: the change in small business revenue in low income zip codes has gone from -24.3% to -30.8%.  It would not be surprising if some of this decline is capturing the effects of the civil unrest following the death of George Floyd.   Somewhere in the neighborhood of 1,500 buildings were damaged throughout the Lake, University and Broadway corridors in Minneapolis and St. Paul, many of them small businesses.   An unknown number of other businesses were undamaged but boarded up and closed.

Tools such as this can offer a better and more refined understanding of public policy problems.    Whether that translates into better public policy solutions will always be an open question.    Inevitably the insights this type of analysis generates will conflict with political agendas and preconceived notions about what to do and how to do it.   No tool has yet been created to resolve conflicts between ideas like economic efficiency and political acceptability.