According to many policymakers, pundits, and others, Maine’s labor market faces twin structural problems. First there is the skills gap, where the skills demanded by firms cannot be supplied by the current work force. While both jobs and workers are available, the problem is the workers do not have the requisite skills for those jobs. Second there is the demographics issue, where there is simply a shortage of workers to fill the jobs available. These two structural issues are similar in nature (demand for labor exceeds supply), and they both have been largely accepted as axiomatic in the state.
The problem with the presentation and acceptance of these twin structural arguments is that there lacks sufficient data and evidence to support them. In fact, the data and evidence suggests that the problems in the Maine labor market are largely cyclical in nature, stemming from a weak recovery following a collapse in consumer demand. For purposes of this post I am going to focus on the labor shortage argument (and will come back to the skills gap argument in a later post), specifically to address the fact that there are conflicting reports about the impact Maine’s shifting demographics will have on the labor force, why focus on structural issues in the labor market ignores the ongoing cyclical issues, and why ignoring those cyclical problems will both rob the state of an opportunity to boost its labor force while also exacerbating the structural problems down the road.
First, we need to get an idea of the problem the labor shortage poses for the state’s economy as articulated by those who claim Maine’s problems are structurally related. The following is from Charles Colgan, quoted by a report issued by the Maine Chamber of Commerce and Maine Development Foundation:
“From 1990 to 2010, the state lost more than 20% of its population of 18-34 year olds. During the next decade nearly 200,000 people will reach retirement age in Maine. If there is no in-migration to the state, the state’s workforce will shrink significantly between now and 2020. Workforce constraints are currently masked by the recession and lack of recovery, and are much more serious in rural places than urban places, but the problem is so large that no part of the state from Portland to Portage will be untouched by the lack of workforce, if nothing is done.”
Similarly, the following is from a Tom Bell piece published in the PPH last fall as part of that publication’s series covering the state’s shifting demographics (my emphasis):
“So over the next two decades, as the baby-boom generation leaves the work force and retires, the number of available workers contributing to the Maine economy will decline,” said Colgan, who has studied Maine’s demographic data for more than 30 years.
For individuals, the labor shortage will be good news because employers will be forced to increase wages, he said. For companies, however, higher labor costs will make it harder for them to compete with firms outside the state and grow.
The only solution is to bring more people to Maine, he said.
“If they don’t show up, Maine’s economy starts shrinking,” he said. “We simply don’t have the people to do that work.”
There has been a lot of discussion among politicians lately about Maine’s “skills gap,” the perceived mismatch between the skills needed by employers and the skills being taught in schools and colleges.
But just revamping the skills of the current population is not enough, said state economist Amanda Rector.
“There’s a people gap,” she said. “There are not enough warm bodies to go around.”
So, does Maine have a labor shortage, or a job shortage, or both? And if it’s a labor shortage, is it here now, if not when will it hit, etc.? In short, there’s no reason to believe that Maine is facing a labor shortage at the present, nor is there strong evidence that a labor shortage is to occur in the near future.
At first blush, one could conclude from the above comments and characterizations that Maine is currently facing a labor shortage (though Colgan has elsewhere stated that the labor shortage is likely not to occur for a few years). However, there is little reason to believe this is the case.
First, the labor force has continued to grow post-recession:
While jobs are well below pre-recession levels:
Moreover, wages* and hours have remained flat, indicating that demand for labor relative to the supply of labor is not growing. Simply stated, labor is like any other good or service, and like any other good or service, when demand for it grows relative to its supply, the cost (wages) will increase. In lieu of increasing wages, should firms be unable to increase wages offered to potential workers, or there just not be an adequate labor pool for firms to draw from and increasing wages will not increase hiring, firms can increase the hours of their current work staff. However, if neither hours nor wages increase, then there is hardly evidence of a labor shortage. This is the case as both are back to pre-recession levels.:
(click images to enlarge)
These two data point suggest that Maine is not currently facing a labor shortage.
So if not now, then when? Labor shortage is a relative term, as you need labor demand to exceed labor supply. You could have growth in both, or contraction in both, and arrive at the same result. Given the concern about Maine’s demographics, it’s fair to assume that the concern is that the labor force will eventually decline, as claimed in the joint report issued by the MCC and MDF. In that report, it is projected that the labor force will decline by 20,000 workers by 2020. First, this represents just 3% of the current labor force. Second, the report arrives at that figure based on the following assumptions:
If current trends do not change, if population growth continues on its projected path, and if participation rates do not change, the civilian workforce will cease growing
entirely and decline by 20,000 by 2020.
So, based on current trends, Maine’s labor force will shrink by some 20,000 workers. Interestingly though, a report issued by the Maine Department of Labor in May, 2013 conflicts with the MCC/MDF report. In the MDOL report, similarly based on current population and participation trends, the state’s labor force is expected to grow by 18,800 workers. What makes the MDOL reports so interesting is that the MCC/MDF report assumes constant labor force participation rates (“LFPR”) and projects a declining labor force, whereas the MDOL assumes a declining LFPR (to 63.2% in 2020, down from 64.3% in 2010, and which currently stands at 65.5%) and projects a growing labor force.
The actual impact of Maine’s shifting demographics on the labor force are not clear from the above two reports; neither in terms of what the impact will be on the labor force, and in turn what that will do to Maine’s economy.
What has been made clear by Colgan, Rector, and others is that the “only solution” to the labor shortage is through worker in-migration. However, the MCC/MDF report, which begins with the Colgan quote that Maine’s cyclical problems are actually masking the structural ones, suggests that the state push non-labor force participants back into the labor pool; in other words, create more demand for that excess labor. Likewise, from the MDOL report:
The participation rates for age groups younger than 55 dropped during the 2000s due to poor economic conditions and other factors. As a result, there were 29,000 fewer labor force participants aged 16 to 54 than there would have been had participation rates remained at 2000 levels.
In other words, if we push the LFPR for the younger age cohorts back to pre-recession levels, we will offset the declines in the labor force projected by the MCC/MDF report. This piece from the MDOL report also highlights two important things. First, the decline in labor force metrics for the prime age working cohort is a sign that Maine’s labor market weakness is still cyclical in nature. Second, the labor market issues are not structural. Whereas the older (55+) and younger (16 – 24) age cohorts are often structurally unemployed (retirement for the former, school for the latter**), the 25 – 54 age cohort, known as the prime age workers, are typically involuntarily unemployed (unemployed because of downturns in the economy, lack of job openings, etc.).
Of course, typically is not always, so there are instances of people in the 25 – 54 cohort being structurally unemployed, but LFPR and employment/population rates (“EPR”) are higher for this group than the other two. Because of that, this cohort is a good (though not perfect) barometer for cyclical problems in a labor market. In Maine, the labor market metrics for the 25 – 54 year old cohort suggests that the soft labor market is still cyclical in nature.
(click image to enlarge)
While the MDOL report suggests the state can boost the labor force by some 29,000 workers by increasing participation, pushing the EPR for the 25 – 54 age cohort 3 percentage points from its current level (which would bring the EPR to a level at the bottom end of the annual EPRs for the 2000 – 2007 period), the state would gain roughly 16,000 workers.
Furthermore, there are just over 40,000 unemployed persons in the state (a figure that does not include those persons who want to work and have dropped out of the labor force). According to the Maine Department of Labor’s website, as of May 16 there were 7,944 job postings in all industries; a ratio of 5 unemployed persons per job opening. This is not indicative of an economy that is short on “warm bodies.” [EDIT: Using the state's job bank is not an entirely accurate barometer of job availability in the state, so it is difficult to accurately peg job openings. Primarily, not all employers use the state's job bank, or other job posting websites such as Indeed.com and JobsinMe.com. Currently, Indeed.com lists 11,392 jobs for the entire state, while JobsinMe.com 12,310 jobs, and absent more comprehensive data, using website job postings are the only surrogates to such comprehensive data to get an idea of job vacancies in Maine.]
Of course there are industries and geographic locations that might have problems drawing workers, but, and I will delve into this topic when I write on the skills gap, this is not indicative of an economy wide labor shortage, and the appropriate response is for firms to increase wages (we are still a market economy). If a firm (or entire industry or economy) offers noncompetitive wages and fails to attract workers, does that mean there is a “labor shortage”?
Structural and cyclical problems in the labor market require different policies. For the former, policies are aimed at retraining or increasing the labor force, whereas the latter requires an increase in the demand for the available labor. As a result, one set of policies will not address both problems, and misdiagnosing the problems in the Maine labor market can cause problem down the road. As Berkeley economist Brad DeLong wrote in 2010:
For nothing converts cyclical unemployment into structural unemployment more certainly than prolonged unemployment.
By any measure, the state’s (and nation’s) economy turned south because of cyclical problems–there was an enormous decline in aggregate demand and the business cycle nose-dived. Clearly the cyclical problem was the initial cause of most of the unemployment in the state. However, as the cyclical problems went untreated, the unemployed remained unemployed for longer and longer. In 2007, there were only 3,000 people unemployed for longer than 27 weeks. By 2009 that number grew to 8,000, and sat at 13,000 in 2012 (the most recent data available at the CWRI website). Moreover, the average duration of unemployment nearly doubled in less than two years, and has remained well above its pre-recession peak, and as a percentage of total unemployed. One reason for this is that although the total number of unemployed persons has declined (slowly) over the past few years, those unemployed for 27+ weeks has remained relatively steady since 2010.***
(click images to enlarge)
As people remain out of the labor force for longer periods of time, their skills begin to erode and they become less hirable as they compete in a slack labor market against more highly qualified applicants. As a result, those initially unemployed for cyclical reasons become unemployed for structural reasons. That is the course Maine’s policymakers could be charting for the current group of long-term unemployed by persisting to focus on shifting demographics and ignoring the cyclically unemployed by assuming “[t]here are not enough warm bodies to go around.”
Maine’s population and labor force are getting older, and that will have implications for the state in the future. Exactly when that will take place and what those implications will are not clear. What is clear is that one-half of the argument that Maine’s labor market and economic problems are somehow structurally related does not hold up to scrutiny. Currently, there is no labor shortage, and even those who argue that Maine’s demographics will cause a labor shortage down the road admit that the state cannot currently absorb a mass in-migration of workers because job creation remains tepid. There is still slack in the labor market, wages and hours remain stagnant, and long-term unemployment persists. It’s difficult to fathom how policymakers expect to draw workers from out-of-state when the current labor market conditions are so bleak. This is not, in fact, a chicken or egg conundrum; the excess labor supply clearly indicates a need for job growth for the current labor force.
* Yes, we can find industries with rising wages, but that is result is too narrow to comport with the phenomena of an economy wide labor shortage. If that were the case, then we would see enough increases in median wages across various industries to pull up the overall wage.
Moreover, the slight bump in wages during the recession is not likely an indication of a labor shortage or other structural issue. Wages are likely to, at least, remain stagnant if not rise during a recession. Why exactly this occurs is up for debate.
** Increases in unemployment or declines in LFPR due to schooling are not said to be cyclically related, but this is not entirely true. For instance, enrollment rates rise when there are economic downturns (see here, here, and here). Likewise, disability claims (too considered a structural problem in a labor market rather than cyclical) have a cyclical component to them (see here and here).
*** While I intend to leave the skills gap argument for a later post, I have to make note of the fact that the rise in the duration of unemployment (as well as other labor market metrics–here and here) coincides with the economic downturn in 2008. Further, the rise in the duration of unemployment occurred quite rapidly rather than rise gradually over time, as would be expected if the unemployment were structural. As economist Edward Lazear wrote in 2012, “[t]he structure of a modern economy does not change that quickly.” In other words, Maine did not get old, become unskilled, become uneducated, and so forth in the period from 2007 to 2009.