Detroit Economic Growth Falls Below Trend Again in December, According to Chicago Fed Index
The Chicago Fed’s Detroit Economic Activity Index (DEAI) was –1.13 in December 2020, down from +0.08 in the previous month. All the monthly index values for 2020 are displayed in figure 1. Also featured are visual representations of the contributions from the index’s four major categories: labor, trade, real estate, and income. An index value greater than zero points to the city of Detroit’s economic activity growing faster than trend, while an index value less than zero points to the city’s economic activity growing slower than trend.
Following the pandemic-induced economic decline in the first quarter of 2020, the DEAI bottomed out in April at –14.47. According to the monthly index readings, Detroit’s economy then rebounded, experiencing modest growth in May (+0.34) and extremely strong growth in June (+5.69) and July (+3.86); growth in the city’s economic activity then slowed markedly in August (+0.70) before turning negative again in September (–1.12). For the remainder of 2020, the index continued on its rollercoaster-like path, rising sharply in October (+3.13), slowing in November (+0.08), and falling below zero again in December (–1.13). For the entire fourth quarter of 2020, the DEAI value was –0.13, down from a very strong +7.34 for the third quarter. And with an annual value of –2.03 for 2020, the DEAI was down considerably from its value of +0.81 for 2019.1
Figure 1. Detroit Economic Activity Index, by category, 2020
Table 1. Contributions to the Detroit Economic Activity Index, 2020:Q4
| Category | December 2020 | November 2020 | October 2020 | 2020:Q4 |
|---|---|---|---|---|
| Labor | –1.15 | +0.07 | +3.14 | –0.05 |
| Trade | +0.02 | 0.00 | –0.01 | –0.05 |
| Real estate | –0.01 | +0.01 | 0.00 | –0.02 |
| Income | 0.00 | 0.00 | 0.00 | –0.01 |
| Index | -1.13 | +0.08 | +3.13 | -0.13 |
Labor’s contribution to the DEAI for the fourth quarter as a whole was –0.05. After starting the quarter strong in October (+3.14), labor’s contribution declined toward trend in November (+0.07) and fell below it in December (–1.15). This pattern is reflected in the month-over-month change in employment, which was positive in October (+8.5%) and November (+1.7%) before turning negative in December (–3.5%). The level of employment in December was down by 7.5% on a year-over-year basis. Detroit’s unemployment rate—after its spike up to a seasonally adjusted value of 46.5% in April—had fallen to 14.9% by October (still well above its average of 8.7% for calendar year 2019). Seeing an increase in job opportunities in the fall, more potential workers returned to the labor force around then; but given that not all of them found a job immediately, the unemployment rate increased to 20.6% in November, even as employment increased. In December, when employment declined, the unemployment rate rose again, finishing the year at 21.3%.2
Trade also made a contribution of –0.05 to the DEAI for the fourth quarter of 2020. Trade’s monthly contribution to the index started the quarter below trend (–0.01 in October), but finished it slightly above trend (+0.02 in December). December’s positive contribution to the DEAI was due to the noteworthy month-over-month increases in both imports (+2.6%) and exports (+3.5%). On a year-over-year basis, imports (+3.2%) improved more than exports (+1.8%) in December. That said, the city’s trade balance remained positive in 2020, with the Detroit port having exported more goods than it imported over the entire year.3
Real estate made a slightly smaller negative contribution (–0.02) to the DEAI for the fourth quarter of 2020. The most significant changes in this category were the increases in the vacancy rates for commercial real estate and in the prices for residential homes. The average vacancy rate for commercial property increased from 4.5% in the third quarter of 2020 to 6.7% in the fourth quarter. This jump is reflected in the negative contributions from this single real estate indicator to the index for October (–0.01), November (–0.01), and December (–0.02). However, these contributions were largely offset in the index by positive contributions from residential home prices in October (+0.02), November (+0.02), and December (+0.01). For 2020, the average inflation-adjusted prices of residential homes and condominiums in Detroit were up 20.4% and 4.5% on a year over-year basis, respectively.4
The contribution to the DEAI from the income category—consisting almost entirely of annual data—does not tend to change significantly on a monthly basis. We will learn considerably more about the impact the pandemic had on this category when the data are updated later this year. Throughout the fourth quarter of 2020, the income category made a very small contribution—which rounded to zero for each of the three months in the quarter, but was slightly negative (–0.01) for the quarter as a whole, according to the DEAI model.
I now turn to the DEAI’s projection forecast for Detroit’s 2020 real per capita income (PCI). Figure 2 presents the historical real PCI data from the American Community Survey (ACS) for 2014 through 2019, as well as the DEAI model’s real PCI predictions for 2016 through 2020. As reported in my December 3, 2020, blog post, Detroit’s real PCI reached $21,044 in 2019, rising 12.2% from 2018 and outperforming the model’s projection. However, even with stronger-than-predicted results in 2019, real PCI for Detroit was still much lower than real PCI for Michigan ($32,892) and for the United States ($35,672)—which were up 2.5% and 3.6% from the previous year, respectively. In 2020, the DEAI model projects that Detroit’s real PCI will decrease slightly to $20,747—a 1.4% reduction from 2019’s actual level. If this projection holds up, this would shift the trajectory of growth in real PCI for the city.
Figure 2. Real per capita income for the city of Detroit
The DEAI model incorporates the gross metropolitan product (GMP) data for the Detroit–Warren–Dearborn metropolitan statistical area (MSA) from the U.S. Bureau of Economic Analysis (BEA) in order to estimate Detroit real gross city product, or GCP.5 Detroit real GCP indexed to its 1997 level is displayed in figure 3. Over the years 2015–17, Detroit real GCP (averaging 3.3%) grew faster than U.S. real gross domestic product, or GDP (averaging 2.4%). However, during the years 2018–19, Detroit’s real GCP growth rate slowed to an average of 2.3%, while the national real GDP growth rate (averaging 2.6%) remained fairly close to what it had been in the previous three years. For 2020, the DEAI model estimates a 4.9% decline in Detroit’s real GCP on a year-over-year basis. This is worse than the 3.5% decline in the nation’s real GDP for 2020, according to the BEA. While most forecasters are predicting that U.S. real GDP will be back to its level of the fourth quarter of 2019 by the middle of this year,6 it is uncertain how long it will take for Detroit real GCP to get back to its previous peak. If the model’s estimate for 2020 is correct, the city’s economy needs to expand by just under 5.2% to return to its 2019 level of real GCP. A 5.2% year-over-year gain in Detroit’s economy would be larger than any increase in the 24 years of information used in the DEAI model’s estimating process. The largest single year-over-year gain for Detroit real GCP in the model’s data was in 2016, when the city’s economy grew by 3.7% and the DEAI had an annual value of 1.19.
Figure 3. Detroit real gross city product (GCP)
So, why is there such a large discrepancy between expected Detroit real GCP growth and U.S. real GDP growth for 2020? One way to explain this is by looking at the changes in employment-related data for the city and the nation. To show how Detroit’s employment situation compared with the nation’s, I present table 2 below. The comparisons are based on annual averages to match the way the year-over-year percent changes in real GCP are calculated. The table shows that according to the U.S. Bureau of Labor Statistics’ Current Population Survey data, annual average employment fell by 10.6% for the city, but 6.2% for the nation. The number of unemployed rose significantly for both Detroit and the nation; but in percentage terms, the city’s jump of 149.6% was far worse than the country’s jump of 115.8%—which was partially due to the reentrants to the city’s labor force. The city’s labor force actually finished the year higher than at the start of 2020, having moved up by 3.5% on a year-over-year basis. However, Detroit's unemployment rate—which had been running about 5.0 percentage points higher than the nation’s in 2019—was up significantly more in 2020: Detroit’s average unemployment rate was 21.0% for 2020, whereas the nation’s was 8.1%.
Table 2. U.S. and Detroit employment-related data, 2020
| Employment | Unemployment | Labor force | Unemployment rate | |
|---|---|---|---|---|
| % change, YoY | % change, YoY | % change, YoY | annual avg., in % | |
| United States | –6.2 | +115.8 | –1.7 | 8.1 |
| Detroit | –10.6 | +149.6 | +3.5 | 21.0 |
As the data I’ve discussed in this post suggest, there remains much uncertainty about when the city’s economy will return to some sense of normalcy. Besides the greater fluctuations in employment and higher volatility in trade, pandemic-related changes in commercial real estate and the demand for office space pose yet another challenge for Detroit, as suggested by the real estate category’s contribution to the DEAI for the final quarter of 2020. At the start of the pandemic, many office workers were forced to work remotely. According to a nationally representative survey of U.S. workers conducted by the Pew Research Center in October 2020, roughly 20% of the respondents who stated that their job responsibilities could largely be completed from home reported that they teleworked all or most of the time before the onset of the pandemic.7 During the pandemic, however, that share increased to 71%. Additionally, 54% of these workers said they want the flexibility to work from home after the pandemic ends. The Pew study indicates that the transition to working from home was fairly seamless for many, though lower-income workers faced more challenges. Such findings have led some business leaders to consider reducing the size of their respective office space footprints. In addition, research and worker feedback like this have also spurred some employers to figure out how they will implement more-flexible work schedule policies when the pandemic is over. So, Detroit may be facing a long-term decline in the demand for its office space, which could lead to falling office rents and commercial property values. Moreover, if there are fewer commuters to the city than in pre-pandemic times, there could be negative impacts on the city’s tax revenues and service-related businesses, including restaurants, retail stores, parking garages, and entertainment and conference venues. These changes would delay the city’s return to its 2019 level of economic activity. Using the DEAI, we at the Chicago Fed will be tracking Detroit’s economy as it emerges from the pandemic throughout 2021, and we will report our findings for each quarter.
The March 2021 DEAI release (covering the first quarter of 2021) will be posted on June 10, 2021. The most recent DEAI results and future release dates can be found on the DEAI page of the Federal Reserve Bank of Chicago website. A copy of the December 2020 DEAI release can be found here.
Notes
1 The quarterly and annual averages reported for the DEAI are not “simple” averages, but instead “triangle” averages of the monthly values. For growth rate measures like the DEAI, the triangle average best approximates a quarterly or annual growth rate from monthly growth rates. It gets its name from the tent-line nature of weights applied to current and past values in the averaging process. A simple average weights current and past values equally, while a triangle average has weights that peak in the recent past and are equally spaced going backward and forward in time from the peak (e.g., Q4-20 growth = (1/3)*Dec-20 + (2/3)*Nov-20 + (1)*Oct-20 + (2/3)*Sep-20 + (1/3)*Aug-20 growth rates).
2 All the labor-related data values referenced in this paragraph have been seasonally adjusted, and they have also been adjusted for breaks in their time series resulting from the decennial censuses.
3 All the trade-related data values referenced in this paragraph have been inflation- and seasonally adjusted.
4 All the real-estate-related data series referenced in this paragraph have been seasonally adjusted.
5 When the GMP data are unavailable, the model projects Detroit GCP using the available data for the DEAI and its historical relationship with GMP for the Detroit–Warren–Dearborn MSA.
6 For instance, according to the Blue Chip Economic Indicators consensus forecast as of February 10, 2021, U.S. real GDP is expected to return its 2019:Q4 level by 2021:Q3.
7 Pew Research Center, 2020, “How the coronavirus outbreak has—and hasn’t—changed the way Americans work,” report, Washington, DC, December 9, available online.
