The Labor Market vs The Stock Market

Contributor Image
Written By
Contributor Image
Written By
Dan Buckley
Dan Buckley is an US-based trader, consultant, and part-time writer with a background in macroeconomics and mathematical finance. He trades and writes about a variety of asset classes, including equities, fixed income, commodities, currencies, and interest rates. As a writer, his goal is to explain trading and finance concepts in levels of detail that could appeal to a range of audiences, from novice traders to those with more experienced backgrounds.

The labor market and stock market are linked and we’ll go through the basic implications of how and why it’s important.

Basic overview

When there is lots of labor market slack – i.e., high unemployment – policymakers have great incentives to stimulate the economy with lower interest rates and other monetary and fiscal easing measures, if necessary.

This generally produces a strong period for asset prices. Even though the economy is still weak late in a recession and the early stages of the expansion, the stock market is generally back in a bull market. The financial economy leads the real economy.

Once you get to the late cycle and there’s little labor market slack, policymakers begin to slow things down, as they begin to fear inflation and financial instability (e.g., asset bubbles, too much credit creation).

Then they begin to take away policy support. So while the real economy tends to be good during this time, it starts to be a bumpier period for stocks.

Eventually they go too far in tightening policy because it’s hard to get the balance right. So there’s a drop in the stock market, followed by a drop in the real economy,

This then reverses when interest rates are cut and/or other monetary and fiscal easing measures are put in place to rectify the imbalance.

For that reason, we can say that the health of the labor market is a contra-indicator on when it’s best to buy stocks.

When there’s high unemployment you can generally feel pretty good about the forward direction of stocks assuming policymakers are skillful and have the power to make the necessary measures to provide support to the economy.

Conversely, when there’s low unemployment, it’s prudent to be more cautious about stocks as policymakers begin withdrawing support.

Plotting it on a graph: Labor market vs stocks

If we go back and look at previous business cycles in the US (which tend to be coordinated with those of other countries), every time the U-3 unemployment rate gets down to around 4 percent, the forward returns of stocks have been relatively low.

Each of the pink lines mark approximately the point at which U-3 unemployment went down to around 4 percent. This is a point at which policymakers increasingly become concerned about inflation.

And of course, 2020 was derailed not because of a contraction in credit brought along by a rise in interest rates to fight inflation – like most recessions – but because of a pandemic that worked through a big drop in incomes (businesses shutdown to slow infection rates) rather than credit.


Labor market vs. Stock market

labor market vs. stock market

(​​Sources: U.S. Bureau of Labor Statistics; Wilshire Associates)


The Labor Market

The labor market follows the business cycle, so it’s a key factor for traders and investors looking at making tactical moves.

Over the long-run, productivity is the main factor that drives economic growth and asset price movements.

This is the process by which we learn more, do more, and invent more. Over time, more learning and information is gained than lost, so productivity generally increases over time and it’s not as volatile as credit and money cycles.

However, in the short-run, the cycles are more important.

The archetypical way stocks fall

When the labor market tightens to a point that produces excess price pressures in an economy, central bankers typically respond by raising interest rates, tapering asset purchases, and other ways of easing monetary stimulus.

Eventually they go too far because of how hard it is managing the late-cycle trade-offs between output and inflation, so there’s a drop in credit.

During the early stage of this credit crunch or liquidity crisis, there are typically the steepest drops in asset prices.

This catches many by surprise because they think the economy is strong and it’s confusing why it’s happening. (“The market’s being irrational.”)

During this phase, the money coming in to debtors via their incomes and new borrowing is not adequate to meet what they owe to their creditors.

Assets have to be sold to meet these obligations. Accordingly, expenditures are cut in order to stay liquid.

People eventually lose their jobs because there’s less economic activity. Aggregate incomes decline during this period. This causes spending to drop, in a self-reinforcing pattern.

Sometimes investors mistake the initial drop in asset prices as a buying opportunity. This is because they assume prices have fallen while earnings are still okay, giving assets better value.

But asset prices are being sold because of the realization (by some) that corporate earnings are about to fall.

When asset prices fall, this reduces the value of collateral and new borrowing. Less borrowing means less spending. One person’s spending is another person’s income, so incomes fall.

As such, the issue from asset prices falling is not only about wealth. By “wealth” it’s meant that financial assets are promises to pay money in the future.

But there is also the adverse feed-through effect into incomes.

Borrowers’ creditworthiness is measured as a function of:

  • the net worths and credit ratios (value of assets, income, and collateral relative to their debts) and
  • the sizes of their income in relation to the size of their debt service obligations

When borrowers’ net worths and incomes fall faster than their debts, they become less creditworthy.

Consequently, lenders are less inclined to lend.

This sets off a chain reaction which leads to less borrowing -> less income -> less spending -> lower asset prices -> lower collateral values -> less creditworthiness, and so on, which goes on in a self-perpetuating way.

Many borrowers end up being shut out from the credit markets entirely.

This is because the entities that can normally afford to lend are suffering losses on their assets and the forward economic landscape is highly uncertain during periods of substantial losses and high volatility in the financial markets.

New lending dries up and capital isn’t available.

This is often true even for companies that are run conservatively financially and for projects with high expected forward returns.

As the prices of risk assets, this causes risk-adjusted returns to rise.

Central bankers and other policymakers react to fix the situation when conditions become intolerable.

In developed markets today, interest rates are very low. So this presents a conundrum over how to ease when there’s that lack of flexibility to cut rates.

When nominal interest rates can’t be lowered much because they’re already at zero or slightly negative, this is where the government steps in as the lender of last resort and effectively has to reassure investors that they will come to the rescue and save the system.

The government – assuming they have a reserve currency and effectively have the power to do so – will guarantee that large amounts of money and credit will be made available to various entities in the economy so they can rely on a quick recovery.

However, this support from policymakers is usually not promised quickly enough. This means there are typically painful losses before this support is provided.

In most financial crises, there are also typically hang-ups and resentment toward certain actors who contributed to the crisis. Taxpayers don’t want their money used to support them and policymakers want them to have to bear the consequences of their choices.

Nonetheless, once it becomes evident that the cost of not providing support is greater than the cost of providing it, public officials inevitably come in to do what they can to support the whole.

Market participants can begin to see light at the end of the tunnel that capital markets will recover, as will lending activity and the overall economy.

Labor markets are generally weak to very weak at this point. But they improve once enough policy support is provided.

The speed at which they recover depends on the extent of the stimulus provided and the pace of the overall recovery of the credit and overall capital markets.

Labor markets post-Covid

Labor shortages in the US may not be temporary, which have material implications for monetary policy, inflation, earnings, and asset markets.

When labor markets “heat up”, it means labor shortages become more common and eventually get to the point they leave certain labor categories with low unemployment rates.

This will create inflationary pressures on wages and tightening labor conditions. These tight labor conditions may lead to out-sized earnings growth in the corporate sector because the real economy is doing well.

But policymakers at the same time may want to push back if inflationary pressure becomes too extreme.

This will also pressure interest rates higher over time because it means there are fewer unemployed people to draw upon in order to meet the demand for workers. It’s these types of forces that contribute to overheating economies or what some economists call “overheated labor markets”.

Low unemployment is good, but at a point the inflation risk causes financial stability concerns.

Unusual labor market dynamics

At the same time, labor market dynamics are somewhat unusual. Lots of workers are quitting their jobs at the same time there are lots of job openings.

In other words, there’s plenty of work but a lack of workers, with many dropping out.

Many used the pandemic to develop a “side hustle” – a personal business or some form of entrepreneurship that means many won’t return to traditional work or aren’t looking for it.

Some see it as temporary due to the lingering effects of virus-related risks, resistance to workplace vaccine mandates, or related to extended unemployment benefits.

So, as labor markets become tight and unemployment rates fall, inflation will rise and interest rates will follow suit.

Other labor market impacts

The labor market is increasingly affected by worker scarcity and the labor dynamics caused by the pandemic’s displacement of labor and the role of entrepreneurship and automation in labor markets.

These developments likely represent a labor market undergoing structural changes that could impact inflation, supply chains and corporate earnings, the pace of monetary policy tightening, and stock market performance.

Traders and investors will want to understand how much of the labor shortage can be absorbed by entrepreneurship and self-employment and what this means for the future of various businesses.

Automation is likely to accelerate. Businesses also have the option of doing more outsourcing because it’s cheaper than hiring domestic/local employees or raising wages.

A record number of companies:

  • are reporting job openings they can’t fill
  • have raised compensation
  • plan to raise pay over the next three months

This has lowered business sentiment because of the impact on margins and the inability to meet their labor needs.

But it also means that labor scarcity might not have a big impact on corporate earnings for those who can navigate this well, understanding that a lot of workers aren’t coming back.

This means that labor markets are somewhat different from recent years because labor is re-allocating toward “newer companies” (often sole proprietorships because individuals are now their own business) rather than traditional ones.

In addition, labor scarcity could increase long-term inflationary pressures on prices and greater uncertainty in inflation could exacerbate the boom-bust pattern for interest rates.


Where have the workers gone?

Why the scarcity of workers?

First, the workforce has contracted during the pandemic for multiple reasons.

At least 200,000 among the 750,000+ Covid-related deaths in the US belonged to the eligible working population, according to estimates by the Centers for Disease Control and Prevention.

Some may also be experiencing lingering impacts/symptoms and not want to return to work full-time.

There’s the factor of accelerated retirements. According to Morgan Stanley, about two million more people exited the workforce due to retirement than demographic trends would have predicted. Many nearing retirement simply decided it was a convenient time to head out, especially with so much uncertainty at the onset of the pandemic.


covid early retirement


Older workers who could least afford to retire early – namely, those with less education and lower incomes and less education – have been more likely to leave the labor market during Covid-19.

What’s not know is whether their retreat from the workforce is temporary or permanent. On one hand, there’s Covid-19 fears. For others, it was essentially a type of forced retirement after the inability to find suitable work.

Moreover, the rising value of stocks, homes, and other types of assets also has enabled a group of more affluent older workers to retire earlier than they originally anticipated.

Another probable cause is that the rise in the number of retirees is because fewer of them are getting back into the workforce relative to what otherwise might be expected. The shift makes it harder for economic policymakers trying to determine which jobs lost during Covid-19 will return and how many workers will be able and willing to fill them.

So far, it isn’t clear if recent retirees will be lured back to work and how many are staying out of the labor market for good.

New forms of leverage

The pandemic has also given workers more leverage in their quest for better wages and benefits.

“White collar” employees are able to work remotely during a crisis. On the other hand, frontline staff face elevated stress and health risks.

Then there’s the category of recreational and service businesses. They face a particularly acute labor shortage, as its workers switch fields. This is fueling a surge of retraining and career repositioning, facilitated by e-learning and professional certification options, as well as more looking toward entrepreneurship and the gig economy.


The Beveridge curve

The Beveridge curve shows the relationship between the unemployment rate and job openings.

While employment-to-population ratios suggest there should be lots and lots of slack in the labor market, which holds down wages, the post-Covid Beveridge curve tells a different story – that the equilibrium rate of unemployment (often called ‘NAIRU’) has gone up.


Pre-Covid Beveridge curve suggests an unemployment rate at minus-2 percent

Pre-pandemic Beveridge curve suggests an unemployment rate at -2 percent

(Source: Macrobond and Nordea)


The pre-Covid relationship between job openings and the unemployment rate implied that we should be seeing a negative unemployment rate.

However, with many workers still afraid of the virus, or having trouble finding child care, trying to switch from the services sector to the goods sector, the unemployment rate has stayed higher.

Lots of workers were also fired due to workplace vaccine mandates. This made worker shortages worse in some regions (and reduces the flexibility of the US labor market) and is inflationary.

Of course, jobless New York healthcare workers (as an example of those subject to vaccine mandates) are of course welcome to states with more laissez-faire policies. But relocating takes time. This boosts frictional unemployment and therefore NAIRU – at least in the short run.

Other labor market dynamics

Worker shortages can accelerate economic growth by prompting higher wages, new business formation and product development, and, over time, help increase labor supply as more workers enter the workforce due to the increased incentive for entrepreneurship and retraining.

If this happens in a fast enough manner it could mitigate the negative effects of labor scarcity on employment trends.

However, if it takes too long or is too slow workers may find themselves underskilled for the jobs they want because their previous skill sets aren’t employable in new roles unless they get retrained.

Those who experience unemployment or underemployment during this time may face negative consequences such as loss of skills or switching industries that will also affect future earnings power.

There are numerous examples. For example, in journalism, as it becomes increasingly digital, more workers in print and distribution will need to be retrained as those skills are less valued.

The labor market will face labor skill and labor availability issues.

An example of this was the labor scarcity that followed the 2008 recession. The US unemployment rate reached about 10 percent in October 2009. That same year, job postings on online job boards increased by about 40 percent as companies felt comfortable hiring again.

Employers began to seek candidates with more experience and higher qualifications for available jobs. Many employers were deepening their talent pools via recruitment strategies such as employee referral programs or enhanced outreach to social media communities where potential applicants represented much of the audience (e.g., LinkedIn).

On top of this, some companies outsourced some labor and had more flexibility due to technological changes like hiring digital workers remotely and freelancing platforms becoming popular and available during this time frame for technical labor.

As labor scarcity increased, labor prices (i.e., wages) began to rise as employers competed for limited labor supplies.

This left many job seekers with higher options and bargaining power than before the recession.


When labor is scarce

When labor is scarce, labor prices increase and companies are incentivized to invest in labor-saving technology such as equipment and software that can handle more jobs.

This will reduce labor costs but has a longer lead time.

For a shorter lead time, companies can offload labor to labor platforms or freelancers so labor costs can decrease without having to worry about whether the technology will be effective or efficient.

So while labor scarcity leads to increased labor costs, companies can adapt because of technological investments or hiring/retaining higher-skilled employees which require less hours and expenses for retraining and sustainment (i.e., education/training).

And as labor markets tighten it becomes more difficult for employers to fulfill customer demands because they don’t have enough employees for the necessary hours needed. The inability to meet demand leads businesses to restructure or reformat their business model via automation, increased overtime work, restructuring job responsibilities, and hiring employees who can cover multiple roles at once (e.g., full-stack developers).

Due to the cyclical nature of labor scarcities and labor shortages, these workers can expect to see their wages rise as they find employment and will tend to have less bargaining power when there’s lots of labor market slack.

Labor’s impact on margins

These labor dynamics are important not only for monetary policy, but also for company earnings and stock markets.

Labor costs are a significant portion of corporate balance sheets – about 55 percent of S&P 500 companies’ expenses.

This is typically three to four times other line items.

Since labor costs can be volatile and contribute significantly to operating margins, how well the business cash flows, and therefore stock market returns, it’s important to pay attention to labor dynamics and how they affect company earnings and stock prices.

For companies, labor costs vary at around +/−1 percent annually due to the business cycle and company-specific developments. During economic booms, there will be pressure to boost wages but also higher work productivity. During recessions, wage budgets are typically slashed but so is revenue.

So labor is a big deal for businesses – especially cyclical ones – and can be an important tool in assessing business health and stock market performance.


Labor share

Labor share is a concept that involves how income is split in an economy between workers and shareholders – basically what percent of total revenues goes to labor (i.e., salaries/wages).

If labor share falls, profits rise because labor costs fall.

The inverse also holds true. If labor share rises, then labor costs rise.

A decrease in labor share usually means that labor costs are decreasing as a percentage of the companies’ expenses which means labor scarcity is increasing, labor markets are tightening, companies are becoming more efficient with their workers, or automation is making labor cheaper through process re-engineering or technology while revenue stays consistent or increases.

In addition, labor share has been decreasing since the early 2000s. This is significant because labor scarcity should increase labor costs which means labor share should be coming back up – but it isn’t due to a variety of factors such as offshore outsourcing and technology replacing labor.

If labor share were to increase back up to historical averages (if labor continues down the same path) then this would mean labor prices will be under pressure, labor markets will tighten even more due to increased competition for employees (i.e., bidding wars if companies want these workers), business margins will decrease later in the business cycle when labor costs start making up a larger portion of total company operational expenses, and stock prices could experience downward pressure or muted returns because of increased labor costs.

Overall, most statistics show that labor shares are decreasing, labor markets are tightening, labor costs are on the rise, labor prices are under pressure (due to labor scarcity), and labor is becoming an increasingly larger portion of corporate operating expenses.

This is why labor dynamics are increasingly important to consider when investing or trading the business cycle – labor costs, labor scarcity, labor market tightness/efficiency, and labor share all provide important information on how cyclical labor markets can be as well as how labor’s contribution to company operational expenses will change over time.


Distributional impacts

In this environment, traders and investors will need to watch wage growth and inflation metrics.

They’ll also need to consider balancing tactical stock selections between capital-intensive and labor-intensive businesses.

Labor-intensive businesses include sectors like healthcare and consumer staples. Capital-intensive businesses include financials.



Labor market dynamics affect company earnings and stock markets directly via labor cost contributions to overall revenue and margins.

This makes it important for investors to understand how these can fluctuate from business cycle to business cycle depending on whether companies are hiring or firing en masse.

Labor scarcity will impact inflation, labor costs, labor availability, labor movements (labor reallocation), labor market structure (entrepreneurship and gig economy versus traditional companies), stock market performance, wage growth rate trends, monetary policy, and provide long-term structural shifts for employment.

Labor scarcity can put pressure on labor prices which are reflected in wages. When labor is scarce, the labor force has better bargaining power that results in higher wages. These conditions could lead to increased labor costs for employers and stock market returns may be lower for those who can’t adequately adjust.

The labor market post-Covid has experienced labor shortages due to the pandemic’s displacement of workers.

As demand presses up against supply, this will lead to higher prices and interest rate increases which could lead to stock market volatility.

As labor shortages occur, it may be good news for workers until businesses implement more automation and digitization in their business model(s) due to an insufficient amount of traditional labor.

The impact on stock market returns will depend on how quickly businesses adapt to these pressures while leveraging available artificial intelligence or other means of digitizing their business models before these methods themselves become obsolete due to technological advancement and competition from other players.