On January 18, 2024, economist Dr. Marci Rossell presented a talk in Shreveport, Louisiana to a local business organization BRF (Building our Region's Future) that addressed what some term "doom loop thinking" related to the economy. Her talk was intended to break that cycle by explaining how much of it stems from misunderstandings of economic problems of the past and their solutions and misapplying those "lessons" to current economic challenges which may have similar symptoms but vastly different causes. She then focused on explaining how local communities and business leaders can leverage more accurate understandings of current problems and possible solutions to more effectively compete against other communities for new jobs rather than face decline. The talk is available here on YouTube:
Marci Rossell - BRF - January 2024This presentation is highly recommended in its own right solely for the insights Rossell draws from historical and current events we all think we understood / understand but don't. However, her insights about the impact of demographics on economics has other crucial implications for the future that her talk did not address that are worth examining, from both an investment and a personal perspective.
The Recursive Nature of Inflation
Rossell made her point via two key examples. The first example involved the impact of energy shocks in the 1970s on inflation versus worldwide shipping shocks during a pandemic in 2020. The second example involved the impact of changing demographics on labor markets and wages and the resulting impacts on inflation.
In the case of energy, oil prices skyrocketed twice in the 1970s, once in 1973 and again in 1978 and triggered consistently high levels of inflation AND unemployment that lasted for years. Of course, now any time energy prices rise, economists and the public immediately worry about such spikes triggering a similar round of long-lasting, high inflation and slowdowns triggering higher unemployment. Rossell explains that thinking is based upon a misunderstanding of what was really happening in the 1970s. She believes a combination of high inflation and high unemployment is unlikely in the present economy.
In the 1970s, the US economy had been operating for decades on vehicles and a transportation system optimized for cheap oil. Cars and trucks were incredibly inefficient (and polluting) and the US was heavily dependent upon IMPORTS of oil. Manufacturers cannot instantly redesign vehicles to be more fuel efficient (it takes at least 3-4 years to design and ramp up production for a new vehicle) so once the shocks hit, NOTHING could be done in the short term to reduce demand so the extra costs rippled immediately through the economy and continued having an effect over multiple years.
In the 1970s, a much larger portion of the workforce was unionized, including workers making those horribly inefficient polluting vehicles. As contracts came up for renewal, the EXPECTATION of continued inflation triggered demands for higher wage rates, which became part of renewed contracts which CEMENTED higher costs for producers and higher prices for consumers in future years, demonstrating the recursive nature of economic expectations and reality.
In contrast, when energy prices spike in the 2020s, the higher efficiency of vehicles lessens the impact, as does the fact that the US is a net EXPORTER of oil. Rossell reminded her audience that the US produces more oil than Saudi Arabia as of 2023 (12.9 million barrels daily for the US versus 9.7 million for Saudi Arabia). ( Editorial aside... The fact that the true cost of this domestic production isn't factored into the price of the oil in terms of environmental damage from fracking water, etc. is a different topic of economic efficiency and regulatory failure.)
Short term price shocks are also far more isolated in their impact on long term wages in the 2020s than they were in the 1970s. Statistics from the Bureau of Labor Statistics show union membership dropping from 20.1% in 1983 (the first year of their data series) to 10.1% in 2023. Statistics further back are not readily available but union membership in the 1950s was nearly 33%. A drop in union membership as a share of the entire workforce means a much smaller share of wages are being set in advance based on predictions of future inflation based on current inflation. Instead, more wages and salaries are adjusted on a yearly basis as statistics become available but a seven percent inflation rate for one year doesn't assure seven percent wage / salary hikes for the next three to five years for most employees. This reduces the damaging recursive feedback cycle in labor markets that can sustain inflation.
Demographics and Unemployment
Rossell also points out that the "stagflation" of the 1970s was driven by demographic factors unique to that decade that absolutely do not apply in the 2020s. Unemployment in the latter 1940s and 1950s was relatively low because the US was redirecting manufacturing capacity back to civilian products curtailed during WWII, ramping up for new demand from the migration to the suburbs and, in part, helping to rebuild Europe. Unemployment was relatively low in the 1960s because America was launching a space race that triggered significant spending and America was escalating a war in Vietnam. By the 1970s however, unemployment rose for two reasons. First, the first third of the 1946-1964 baby boom generation was reaching working age. Second, the share of women in that generation (and prior generations) wanting to work or needing to work also jumped significantly. In other words, there was a square wave JUMP in labor SUPPLY reaching the economy at exactly the same time an energy supply REDUCTION in a horribly energy-inefficient economy was causing an economic contraction.
The result of this "stagflation" and the political backlash applied by voters to politicians cemented a "truth" in the minds of policy makers that inflation must be crushed before it gets too high and triggers high unemployment. But is that universally true across all time or was it only "locally true" in the specific context of the American economy of the 1970s?
Rossell argues the latter, stating that demographics make it almost impossible for a given level of economic decline now to trigger 1970s levels of unemployment. Why? When analyzed in simplified fifteen year generational blocks, Generation Z (1997-2012) had a total of 69.58 million people who began hitting the work force at age 18 in 2015. The Millennial Generation (1981-1996) preceding it had a total of 72.24 million people who began entering the workforce in 1999. That's a drop in the theoretical upper bound of the labor pool in that generation of 2.66 million workers.
To the extent demographics are economic destiny, extrapolating yearly births to fill out the rest of Generation Alpha (2013-2027) using a steady birth figure of 3.66 million for 2022 through 2027 means Generation Alpha will only total 56.48 million. As that generation reaches working age beginning in 2031, it will contain 13.1 million fewer possible workers than the generation that preceded it. The total US labor force is estimated to be around 163.8 million in 2024 so a reduction of 13.1 million is an eight percent reduction in supply.
If you want to stare at the charts and what-if the numbers yourself, take a look at these two graphs provided on Statista's web site:
According to simple concepts of supply and demand, even if the overall economy itself stays flat over the next ten years, a SHRINKING number of yearly births creates a guaranteed reduction in the labor supply with an eighteen year lag. For a given level of technology, capital and labor in the economy, a shrinking labor pool will inevitably produce a tighter labor market and lower unemployment.
Matching Demographics to Current Reality
How does a demographic centric analysis of labor markets and unemployment match current reality? Pretty tightly, it seems. Economists studying labor markets have theorized there is some "natural rate of unemployment" in an economy which reflects an equilibrium point at which the number of people unemployed but looking for work is solely due to a) a percentage of the work force that voluntarily quit a job to look for a better job and b) some remaining percentage of the work force that inefficiencies in the job market cannot line up with available work in an economy that is growing exactly in sync with population. According to legacy economic theory, most economists estimated the natural rate of unemployment to be around 4.5 to 5.5 percent.
Of course, unemployment in the US has averaged under 4.0% from 2018 to the present except for the period between April 2020 and October 2020 where it spiked to 14% due to COVID lockdowns then returned to 6.7% then back down below 4.0% by December 2021. Most economists, politicians and citizens would have NEVER expected unemployment to snap back to historically low rates in months after spiking to levels not seen since the early 1980s. This persistent low unemployment is a source of concern for many economists looking at 2020s America through the biases learned in 1970s economic textbooks. When analyzed from a demographic perspective, the low unemployment makes PERFECT sense and has nothing to do with over-stimulation of the economy through deficit spending. An economy of a certain dollar aggregate size serving a certain number of people with a shrinking number of people of working age in the work force is going to have lower levels of unemployment. As Rossell stated in her talk, it's tempting to blame perceived labor shortages on lazy teens and twenty-somethings who are unwilling to work. It's a simple theory to explain. It's sound byte friendly. It might serve the political purposes of some. But it is absolutely not the correct explanation of the problem. It's not that they are unwilling, THEY WERE NEVER BORN.
Mapping Demographics to Current Policy
This enlightened understanding of the relationships between demographics, labor markets and unemployment might be an interesting topic for discussion at the monthly Economics Department social at Whatsamatta U or the after-party for the next Federal Reserve Open Market Committee meeting but what are the immediate policy implications? Two obvious implications come to mind, one economically related to monetary policy and attempts to "control" unemployment through interest rates and another politically related to immigration policy.
In the monetary realm, traditional thinking might associate historically low unemployment with an over-stimulated economy trying to grow faster than population growth. Old thinking would immediately trigger a suggestion to raise interest rates to slow down economic growth, slow demand for workers and RAISE unemployment to a more "natural" level. Put a little fear into worker's eyes so they aren't so greedy and drive up wage rates and trigger inflation. In reality, purposely lowering immediate demand for workers does nothing to CREATE more workers for a larger denominator in the unemployment calculation. In fact, boosting interest rates makes it more expensive (and risky) to make capital investments that can substitute automation for human labor, reducing productivity gains and contributing to continued demand outstripping that shrinking available labor pool.
Politically, an enlightened understanding of current labor market dynamics would emphasize the fact that "domestic" supplies of labor inherently involve an eighteen year lag from any change in potential workers entering the pipeline (being born) and entering the eligible work force eighteen years later. That means the only IMMEDIATE variable that can be altered to provide an increase in available workers involves immigration and drawing in supply from other economies via work visas, green cards and permanent citizenship. Of course, that understanding runs immediately into the current buzz saw regarding demagoguery over immigration and "foreigners coming to steal American jobs."
Again, the reality is that such immigrants would not be coming to take jobs that would have gone to Americans. They would be coming to fill current demand for workers that current domestic supply cannot fill, even at $15.00 or $18.00 minimum wage rates. In this scenario, the problem is predominately driven by the SUPPLY of labor, not its PRICE. It is this dynamic that merits more consideration for the future, both from a purely economic theory standpoint and an individual standpoint.
Demographics and Capitalism
If the mathematics of the relationship between downward demographic trends in births and labor supply make sense, it is equally important to consider how that dynamic has shaped capitalism as practiced in most industrial economies for the last two hundred years. Generally defined, capitalism is a collection of economic and legal constructs operated together to encourage economic growth by organizing activities to reward private ownership of capital and individual ownership of labor to yield an optimal balance between capital and labor that produces the largest possible profit. Owners of capital gain by capturing the profits and labor gains by earning wages in an economy with an increasing range and quantity of things to consume.
Capitalist economies come with some known "genetic" flaws that manifest themselves in situations involving natural monopolies, anti-competitive collusion to create "un-natural monopolies" and externalized costs which distort market pricing to induce too much of a good or service by failing to account for all of its true costs. Guarding against these inherent flaws requires appropriate anti-trust regulation to avoid undue consolidation in an industry to avoid outright monopolies or oligopolies where a handful of "competitors" serve an entire market.
Of course, many countries have NOT aggressively enforced antitrust laws in the past 20-40 years and oligopolies of 5-7 key competitors seem to be the norm in most of the most notable businesses (profitable or not) formed during the modern digital era. Many of these new industries didn't start with only 5-7 competitors. There may have been dozens initially but firms attempted to grow market share rapidly by growing sales rapidly. This has been especially prevalent in service industry sectors where entrepreneurs formed companies aimed at leveraging technology to disintermediate existing service business models, somehow make them more efficient and extract profits by doing so. Attempts with this strategy are still pending in service functions ranging from taxis, package delivery, food delivery and even shopping.
Think for a moment what these firms are actually doing. There's SOME traditional "capital" involved, typically in the form of servers and cloud computing bills and some software development costs. However, the core of the business model simply involves subdividing labor that normally would have been done by three or four existing people in the industry (e.g. a matre'd, a waiter, a chef) and added another worker or two to the equation (the packer in the restaurant, the driver) or took labor previously performed by a customer (pushing a grocery cart through a store or retail shop) and replaced it with more workers doing more work at even simpler layers of complexity (a warehouse picker, a packer, a driver).
Both of these examples -- a food delivery app and a grocery / retail ordering app -- are using technology to "dumb down" common tasks into even simpler tasks that can be farmed out to low-skill labor with low wages to squeeze costs in those layers and drive whatever savings can be squeezed to the top. This isn't a new innovation. The term McJob entered the US lexicon in the 1980s as a term for low-skill, low-paying jobs that offer virtually zero room for advancement. In the 2010s, a newer term became popular as new forms of work became "automated" and disintermediated via these technologies -- the gig economy.
Like central bankers attempting to solve 2020s inflation with 1970s concept of macroeconomic theory about inflation and unemployment, capitalism itself has evolved over decades to assume this approach of further subdividing existing required labor into an ever larger number of simpler, dumber tasks and bidding that work out to an ever-growing available labor pool will always be viable. When the population is growing, there will always be a growing number at the bottom of the skills pyramid to perform more and more scut work.
But this analysis has just pointed out that assumptions of an ever-expanding labor pool are decidedly INCORRECT. At least over the next ten to fifteen years based on current birth rates. What are the implications for these business models? NOT GOOD. But... but... but... What about the emergence of artificial intelligence? Isn't that bound to completely automate and eliminate vast swaths of current "professional" work and "content creator" work like musicians, copy writers, reporters and bloggers? Won't that "free up" thousands or millions of workers to slide DOWN the skill pyramid and compete for these McJob / gig work positions? Maybe.
In general though, it seems likely that any business model focusing on squeezing efficiencies out of the "last mile" of the delivery chain of a product or service will encounter labor shortages and labor cost spikes that will make a strategy of substitution TOWARDS labor very risky for the business owners and investors. Even in a more traditional manufacturing model, a company operating at a scale to produce 100,000 widgets employing 500 workers at $39/hour might try to redesign the manufacturing process into more tasks that might require 650 workers but only at $23/hour for the same number of hours. That process works GREAT for improving profits that float to the top as long as there are 150 extra workers to be hired. The availability of those extra workers willing to work for less has been a given for seventy years, from the 1950s to 2020. That availability is no longer a given, AT ANY WAGE RATE. The bodies simply weren't born to be available in the work force.
Demographics and the Individual
Pondering the effects of demographics, labor markets, capitalism and business leads to some potentially unsettling implications for individuals as well at the family level. Any time the birth rate per woman for a country drops below 2.0, overall population growth drops below "replacement level" and population will decline. In a world where medicine has become highly effective at mitigating systemic causes of death at earlier ages, more people will live into their seventies and eighties where diseases of old age crop up and cannot be cured, only treated and mitigated - often generating astronomical costs and unmeasurable stress for family members.
A world with increasing life spans but declining birth rates creates a reverse demographic pyramid which will subject a larger share of citizens to much higher levels of economic and emotional stress caring for aging parents for a longer period of time. The advent of artificial intelligence does nothing to improve the labor situation in nursing homes, "memory care" facilities, etc. An AI system might be able to detect a parent hasn't taken required meds but an AI system cannot enter the room and physically ensure the meds are taken. An AI system will not be able to pick up a parent after a fall. An AI system will not be able to assist a bed-ridden patient with using the toilet, turning the patient or providing a bath. These tasks will ALWAYS require labor.
What if that labor isn't available? Wage rates can rise to encourage people to give up other low-end jobs but at some point, that cost will appear in monthly costs charged to families, pricing some families out of nursing homes and similar facilities.
What about paying for in-home recurring care? Providing lower acuity care in the parent's home might defer demand for more intensive care with more labor later but still puts pressure on labor markets. And in-home hourly help is expensive and limited in the care that can be provided.
So the kids can take care of the parents right? Well, in a family with three or four kids, all located in the same city as the parents, maybe that works and results in a tolerable division of labor. What about a family with only two kids? That's two kids to handle what might be two distinct 2-4 year periods of rising acuity and stress when even one such stint is enough to impact work and care of grandkids. What about only children? Attempting to take care of two aging parents with potentially acute health problems alone would be virtually impossible.
There's no easy answer to this type of problem -- one that is inherently macroeconomic and microeconomic at the same time. What's worse is few current public discussions even reflect an understanding of the problem, much less a hunt for solutions.
WTH
