Date Published: November 9, 2020
One of the most striking aspects of the COVID-19 downturn is the sharp and massive disruption it dealt to the U.S. labor market. A swift deterioration in labor market conditions at the onset of the crisis was soon followed by a sharp (though unmatched) rebound. But, with many of the easy gains now behind us, the labor market recovery has transitioned to a more moderate pace. Like the virus’ path, the initial labor market impact and subsequent recovery have been far from uniform across the country. In this report, we compare and contrast how East Coast state job markets have fared so far during the pandemic. Such analysis requires looking beyond headline indicators. For example, a sharp decline in the official unemployment rate can be misleading if the main thrust behind it is a pullback in the labor force – a common trend during the pandemic.
Note that while national data is available for October, state employment figures are delayed, with data extending to September used for cross-state comparisons. Experiences vary across states, but the broad strokes are that the Northeast continues to lag in the recovery of payrolls, despite some good progress since May. Payrolls in the southern half of the East Coast are generally closer to their pre-pandemic levels. But, since these states lost relatively fewer jobs and have less catching up to do, job growth in this part of the region has already shifted into low gear. Alternative measures of labor underutilization suggest that states like Massachusetts, Rhode Island, New Hampshire, New York and Florida continue to record the largest slack relative to pre-pandemic levels. Faring better are states like Maine, Connecticut, Pennsylvania, D.C., West Virginia and South Carolina. Note that this is not to be confused with direct comparisons of current labor market slack, where pre-crisis underperformers (i.e., D.C., West Virginia) still rank poorly.
While new COVID cases per capita across most East Coast states are still lower than the U.S. average, the majority are confronting a renewed up-wave. A further increase in restrictions would put a near-term damper on recoveries, especially if no additional federal stimulus is implemented in short order. This would be particularly problematic for states that are generally further behind in their recovery and reopening processes.
The East Coast continues to fare slightly worse that the rest of the country when it comes to getting back to pre-pandemic levels of employment. Part of the reason for this is that the region was hit harder than the rest of the U.S. at the onset of the health crisis. Despite some progress in recent months that has helped narrow the gap, the region’s payrolls are still down 7.7% from the pre-pandemic level, compared to 6.7% for the rest of the country (Chart 1).
That said, the southern half of the East Coast is generally doing much better than the northern half. Slicing the region further into four pieces, reveals that the Lower South Atlantic is faring best, with the level of employment there down only 5% relative to the pre-pandemic level – a better turnout than the U.S. as a whole (Chart 2). At the other end of the spectrum is the Middle Atlantic region, where payrolls are still down 10%.
The divergence among the subregions, and across states, can generally be broken down into the severity and evolution of the health crisis, the containment measures put in place and the level of exposure to the most-affected industries. The northern half of the East Coast recorded a sharper spike in COVID-19 cases early on, with the trend more pronounced in the Mid-Atlantic states of New York and New Jersey (Chart 3). This led to an earlier pause in non-essential businesses and stricter containment measures, weighing on payrolls (Chart 4). What’s more, further north, the labor markets of several of the smaller New England states were more reliant on tourism-related industries, such as in the leisure and hospitality, which has borne the brunt of the pandemic and weighed on the overall employment recovery – elements that help explain why the Northeast has lagged behind.
An earlier and more pronounced hit to the labor market would certainly be a disadvantage when assessing where the level of employment stands now relative to the pre-pandemic level. But, another way to look at the recovery is to examine what share of the lost jobs each state has recouped so far. Said differently, it may not be about how hard you fall, but how quickly you can get back up. From this point of view, we can see that except for the large states of Massachusetts and New York, all other Northeast states have recovered a larger share of lost jobs relative to the U.S. (Chart 5). Losing more jobs early on meant that Northeast states had more catching up to do. As a result, job growth here has continued to run at a stronger pace relative to the U.S. average (Chart 6). Success in suppressing the virus’ spread over the summer also played a part in facilitating, or at least not impeding, the recovery process across most Northeast states.
The experience in the southern half is more mixed on this front, with the Upper South Atlantic trailing behind in recouping jobs lost in the March-April period. In this sub-region, the District of Columbia (D.C.) stands out as the jurisdiction that has recovered the lowest proportion of lost jobs at just one fifth so far. D.C. is unique in the sense that the decline in payrolls at the onset of the pandemic was more protracted (spanned three months instead of two) and the fact that job growth has recently turned negative again, at a time when the vast majority of states have continued to rebuild their payroll tallies.
Given that the southern half lost relatively fewer jobs at the onset of the crisis, it had less catching up to do, and job growth in the region has already shifted into much lower gear. Monthly gains have averaged a little over 0.5% in the three months ending in September – about half the Northeast rate. Contributing to this trend is the fact that the virus’ spread in southern region trended higher throughout the summer. While a more elevated virus spread may not be directly tied to more restrictive containment measures, this element still poses an obstacle to normal business operations and plays a part in restricting payroll growth via reduced confidence (both for consumers and businesses) and resulting impacts on spending, particularly in high-touch consumer-related industries.
Tourism and consumer-related industries were among the hardest hit areas of the economy as the pandemic struck. Despite some consistent progress across most states on this front, these areas continue to bear the brunt of the pandemic. This trend is evident in the leisure and hospitality industry, where the level of payrolls is still down by double digits in percent terms nationwide and across all East Coast states (Table 1). The comeback of this industry is likely to be more gradual given the containment measures that are still in place across many jurisdictions (i.e. limits on large gatherings, capacity constraints on retail shopping, restrictions on indoor dining etc.) and reduced mobility, both domestic and international, in light of infection concerns, border closures and state-by-state quarantine requirements. A rising infection spread across most states poses additional downside risk for the industry, given the potential for more restrictions and a further reduction in mobility. To this end, several states have recently expanded their quarantine advisory lists by adding more jurisdictions. Conversely, the potential for an increase in rapid-testing capabilities could lend a helping hand.
However, the health-induced crisis has spread its tentacles on many other parts of the labor market. Apart from the federal government, payrolls in most categories remain below their pre-pandemic level across East Coast states (Table 1). It’s worth noting that the strength in federal payrolls reflects an earlier spike in hiring for Census 2020 positions – a support element that has already seen some rapid unwinding in the last two months as these temporary jobs end. Among private industries, performing somewhat better is finance and insurance, with the level of employment across most states for this industry either very close to, or already above pre-pandemic levels. Professional and technical services and retail have also shown more resilience than average, with the level of employment in both industries a little over 4% below their February levels. However, the Mid-Atlantic region – and New York in particular – is an outlier to this latter theme, given that it is faring much worse in both categories.
|Table 1: Change in Payrolls by Industry|
% Change in Payrolls
|US||East Coast||US ex-EC||New England||Mid Atlantic||Upper South Atlantic||Lower S. Atl.|
|Goods producing||Mining/Natural Res.||-13||-9||-14||0||-27||0||-10||50||-13||0||-20||-12||-8||-4||-9||-10||-2||-2||-7|
|Trade, Transp. & Utilities||Wholesale||-5||-5||-5||0||-9||-3||-8||-4||0||-4||-11||-6||4||-9||-4||-2||-2||-8||-3||-3||-10|
|Financial activities||Finance & insurance||0||-1||0||-2||-1||-6||0||-1||-3||-2||-3||-1||1||-2||2||1||-2||-2||2||3||1|
|Real estate & Leasing||-7||-8||-6||-12||-7||-4||7||3||-10||-12||-8||0||-13||-15||-10||-8||-10||-19||-7||-12||-3|
|Professional and business services||Professional & tech.||-3||-4||-3||-3||-2||-3||-5||-5||-6||-8||-7||-7||0||-13||-3||-2||-2||-4||-2||-4||-5|
|Management & Admin||-9||-9||-9||-6||-8||-5||-10||-10||-3||-10||-17||-7||-20||-9||-11||-4||-9||-10||-8||-3||-3|
|Education & Healthcare||Educational services||-9||-9||-9||-8||-8||-15||-1||-3||-11||-13||-11||-15||-11||-13||-4||-8||-5||-11||-8||-3||-9|
|Leisure & Hospitality**||Arts/Ent./Recreation||-31||-31||-32||-20||-46||-37||-17||-28||-2||-44||-42||-21||-55||-46||-30||-30||-43||0||-21||-26||-22|
|Accomm. & food serv.||-21||-25||-19||-19||-35||-26||-24||-23||-39||-25||-37||-27||-41||-20||-18||-25||-18||-19||-20||-13||-20|
|Other Services||Other Services||-8||-9||-7||-13||-20||-10||-3||-14||-7||-10||-14||-13||-3||-14||-7||-3||-4||-2||-9||-6||12|
|*Detail for some of the smaller states/D.C. may not be available. **Has been omitted from color scheme due to widespread double-digit declines. Source: BLS, TD Economics. October data is available for U.S. only. For consistency, data extending to September is used for cross-state comparisons.|
Payrolls in all other remaining categories are still down between 5% and 10% from their pre-virus peak across the combined East Coast region. While there are some positive outliers with respect to this trend – i.e. New Hampshire’s real estate and leasing employment has surpassed the pre-pandemic level – these are few and far between. On the other hand, negative outliers are more common. These include Vermont’s construction sector (-24%), Massachusetts transportation and warehousing (-20%), Delaware and West Virginia’s real estate and leasing (-15% and -19% respectively), New York’s management and administrative services (-17%), or Maine’s and Pennsylvania’s educational services (-15% apiece).
While deviations from the pre-pandemic level of employment help shed light into the individual industry pain points, it is the level of exposure to these impacted industries that determines the overall drag on a state’s payroll recovery. By calculating the respective contributions to overall changes (i.e., the weight of each industry times the size of its decline/increase), we note that the steep contraction in the leisure and hospitality industry remains the biggest drag across states. This drag is particularly pronounced in D.C., Vermont, Massachusetts and New York (Chart 7).
Removing the impact from the badly-bruised leisure and hospitality sector brings most states’ labor markets closer to their pre-pandemic level, with South Carolina and Georgia roughly just 2% lower. That said, the level of employment across most states remains down substantially even when removing this hard-hit area. This means that the negative impact of the health crisis is deeply ingrained in the wider economy, particularly for states like New York, Delaware, Massachusetts and New Jersey (Chart 7).
The monthly official unemployment rates generated by the BLS typically provide a good read as to the level of slack in the labor market. That said, these figures can be skewed during non-typical times, with the COVID-19 pandemic a prime example.
When we think about a declining unemployment rate, the usual assumption is that more workers have found jobs, thus changing their status from jobless to employed. But, this is not the only way. Another factor that can drive down the jobless rate is a drop in the labor force participation rate.
The pandemic has changed the way people engage with the labor market, generally by reducing participation. The U.S. labor force participation rate fell 3.2 percentage points between February and April. Despite some improvement in recent months, it remains well below its pre-pandemic level. Several factors are contributing to this trend. These likely include the perceived dangers of being infected by COVID-19 while on the job and the fact that job prospects among some industries may be very low at the moment (i.e., accommodation and food services), leading some workers to not pursue employment opportunities. The need to take care of children due to closed schools or kindergartens may also make it difficult for some parents to engage in the labor market. The latter seems to be an important influence behind the outsized decline in in the female labor force participation rate so far into the pandemic. Those that do not look for work (whether they want a job of not) are not counted either as unemployed or being part of the labor force, which in turn can shrink the unemployment rate. This theme of a reduced labor force participation rate (and ultimately, a smaller labor force) relative to pre-pandemic levels is featured across most East Coast states – in some more than others – thereby sending mixed signals about the progression of the recovery (Chart 8).
A crude way to tease out a fuller impact of the pandemic may be to look at what unemployment rates would look like if the level of engagement in the labor market hadn’t changed. To do this, we keep the labor force constant at its February level and calculate a ‘counterfactual’ unemployment rate based on existing employment data from the household survey. Calculating the unemployment rates this way means that, with few exceptions, rates are much higher than what the official figures imply, and that progress has generally been much slower (Chart 9). The biggest differences between (lower) actual unemployment rates and (higher) counterfactual rates are in states like Maryland, Vermont and New Hampshire – the same states where pullbacks in the labor force participation rate are more pronounced.
Based on the makeup of their population or structure of their economies, some states can have inherently higher or lower unemployment rates relative to the U.S. average. For example, states with older populations and thus low participation rates, including several in the Northeast (i.e., Vermont, Maine), tend to have lower unemployment rates. So, instead of simply looking at the level of unemployment rates, a more fruitful exercise is to look at how high rates are now compared to their pre-pandemic levels. In Chart 10, we outline pre vs. post-pandemic ratios for both official and counterfactual unemployment rates. This chart suggests that Massachusetts, for example, ranks poorly among East Coast states, with the official and counterfactual September unemployment rates still respectively three and a half and four times their February levels. Others that fall into this camp by ranking above the U.S. average (i.e. faring worse) on both ratios are Rhode Island, Florida, New York, New Hampshire and Virginia. At the other end of the spectrum, faring better are Maine, Connecticut, Pennsylvania, D.C., West Virginia, South Carolina and Delaware.
Another data piece that can help shed additional light into the matter is the “U6” unemployment rate. This is a wider jobless measure that factors in not only the officially unemployed, but also marginally attached workers and those employed part-time for economic reasons. This tally is then squared against a wider measure of the labor force, which encompasses marginally attached workers. The ‘U6’ rate tends to paint a less rosy picture than the official unemployment rate. In October, the U6 national rate stood at 12.1% versus 6.9% for the official rate.
At the state level, this data has some limitations, such as the fact that it is only available on a four-quarter moving average basis, which may blunt some of the precision we crave during these turbulent times. Nonetheless, its wider reach and consistency make it worth exploring when looking at overperformers and underperformers. Mapping out pre-pandemic versus current U6 unemployment rate ratios alongside the counterfactual unemployment rate ratios, we notice many of usual suspects reemerge in Chart 11, with Massachusetts, New Hampshire, Rhode Island, New York and Florida faring worse. Vermont is now also added to the list. Meanwhile, states where the current level of slack relative to pre-pandemic levels is below the U.S. average include Maine, Connecticut, Pennsylvania in the north, and D.C., West Virginia and South Carolina in the south.
The path of the recovery will ultimately depend on that of the virus and the troubling trend in new infections is cause for concern. While the level of new infections per population remains below the U.S. average in nearly all East Coast states, it is still moving higher across the board (Chart 12).
So far, instead of wide-reaching restrictions to stem the spread of the virus, governments have generally opted to target high-risk areas with more precision, both from a geographic and industry perspective. This helps minimize the economic disruption. The small state of Rhode Island, which is recording the fastest spread in new cases, recently reduced the size of social gatherings from 15 to 10 people and banned spectators from sporting events. In Massachusetts, more than two dozen communities have recently had to take a step back in their reopening process, which entails the closure of some nonessential businesses, while a 9:30 PM curfew has been implemented for gatherings at event venues and in-person dining. Connecticut imposed a similar curfew, while also reducing restaurant capacity back to 50% from 75%.
While hospitalizations are rising, the fact that death rates remain relatively low compared to earlier in the cycle, and that there’s generally still some room before bumping up against healthcare capacity constraints, appear to be suppressing some of the urgency to implement more restrictive measures. Nonetheless, recent developments in Europe, where the closure of non-essential businesses and lockdowns have made a comeback, make it clear that wider-reaching restrictive measures are not that far-fetched if the virus’ spread is not brought under control. A return of more restrictive measures, at a time when the economy is losing steam, would risk stalling or potentially even sending the recovery process of some East Coast states into reverse, especially if no new additional fiscal stimulus is received soon. Intuitively, states that are further behind in their recovery process would be at a greater risk.
Just like the virus’ path, the initial economic hit from the health crisis and the recovery process that followed has varied across East Coast states, with some doing better than others. The Northeast region continues to lag in the recovery of non-farm payrolls, given a more pronounced hit from the pandemic early on. However, with more catching up to do, job growth here has continued to run at a higher gear. On the other hand, payrolls are closer to their pre-pandemic level in the southern half of the East Coast. That said, job growth in this part of the region has already shifted into much lower gear.
Official unemployment figures appear to underestimate the crisis’ impact and overstate progress in the recovery. Alternative measures of labor underutilization suggest that some of the largest states in the region, namely Massachusetts, New York and Florida, along with the smaller New England states of New Hampshire and Rhode Island are faring worse when comparing the current level labor market slack to pre-pandemic levels. On the other hand, Maine, Connecticut and Pennsylvania in the north, and D.C., West Virginia and South Carolina in the south are faring relatively better on this front.
The recovery’s path is ultimately tied to that of the virus. With the virus’ spread now at a record high nationally and new infections rising along the East Coast, the recovery risks losing even more steam given the increased potential for more restrictive measures. The latter would be more problematic for states further behind in the recovery curve, with states like Massachusetts and New York unfortunately falling into this camp.
This report is provided by TD Economics. It is for informational and educational purposes only as of the date of writing, and may not be appropriate for other purposes. The views and opinions expressed may change at any time based on market or other conditions and may not come to pass. This material is not intended to be relied upon as investment advice or recommendations, does not constitute a solicitation to buy or sell securities and should not be considered specific legal, investment or tax advice. The report does not provide material information about the business and affairs of TD Bank Group and the members of TD Economics are not spokespersons for TD Bank Group with respect to its business and affairs. The information contained in this report has been drawn from sources believed to be reliable, but is not guaranteed to be accurate or complete. This report contains economic analysis and views, including about future economic and financial markets performance. These are based on certain assumptions and other factors, and are subject to inherent risks and uncertainties. The actual outcome may be materially different. The Toronto-Dominion Bank and its affiliates and related entities that comprise the TD Bank Group are not liable for any errors or omissions in the information, analysis or views contained in this report, or for any loss or damage suffered.