Who Cashed Our Productivity Paychecks?

Does more labor productivity raise people’s living standards? The conventional wisdom is still “What’re you kidding? Of course it does!” But the evidence on that is pretty sketchy and has been for a while now. So, let’s do a little myth-busting as we explore the so-called productivity-pay gap.

Investopedia nicely sums up the standard line on productivity: “The level of productivity is the most fundamental and important factor determining the standard of living. Raising it allows people to get what they want faster or get more in the same amount of time. Supply rises with productivity, which decreases real prices and increases real wages.”

You can find the same basic claim all over the place, from the The Library of Economics and Liberty to McKinsey to Forbes.

Just one little problem, of course. The data indicates it’s not true, at least not in the ways it has usually been explained.

We’re A Lot More Productive, But Not Much Richer

In the U.S., productivity has been going up for many years. In fact, it rose a little faster between 2019 and 2022 than it did the previous 12 years. Have a look at this data from the Bureau of Labor Statistics (BLS):

Productivity Change in the Nonfarm Business Sector, 1947-2022

Productivity hasn’t grown as quickly over the last 15 years as it had the previous 17. But from 2019 to 2022, it was still growing at a similar rate as it did from 1973 to 1990. Overall, despite an occasional dip here and there, there’s been steady growth.

Sure, there’s plenty of room for economists to complain, but consider the fact that labor productivity more than doubled between 1979 and 2022!

So, if it’s true that “productivity is the most fundamental and important factor determining the standard of living,” then surely our standard of living also doubled in that same time period, right?

The Productivity-Pay Gap

Well, no, not by a long shot. But the answer requires more nuance than that. After all, there’s no clear definition of “standard of living” and productivity itself comes in various flavors. Let’s stick with labor productivity, which compares growth in output to the growth in hours worked, and let’s use inflation-controlled compensation as a more measurable version of standard of living.

Here’s what we get, according to the Economic Policy Institute:

The idea, of course, is that productivity and compensation rose pretty much in parallel up until the early 1980s and then split off from one another. In fact, productivity rose 3.7 times faster!

Which suggests that something’s wrong with the whole conceit and with the fact that so many trusted sources keep claiming they rise in virtual tandem despite solid evidence to the contrary.

How Do We Explain What Happened?

So, how can we explain the productivity-pay gap? There are various theories, but here are three that, while not necessarily contradictory, stress different facets of the gap.

Theory 1: Policymakers Tore Out the Coupling

The EPI itself, which has a somewhat left-leaning orientation, explains it like this: “Starting in the late 1970s policymakers began dismantling all the policy bulwarks helping to ensure that typical workers’ wages grew with productivity. Excess unemployment was tolerated to keep any chance of inflation in check. Raises in the federal minimum wage became smaller and rarer. Labor law failed to keep pace with growing employer hostility toward unions. Tax rates on top incomes were lowered. And anti-worker deregulatory pushes—from the deregulation of the trucking and airline industries to the retreat of anti-trust policy to the dismantling of financial regulations and more—succeeded again and again.”

In other words, the government allowed the system to get misaligned. Let’s use the metaphor of a coupling. In machinery, a coupling is a device for joining two rotating shafts at their ends so as to transmit torque from one to the other. The goal, of course, is to transmit power fairly evenly. In the coupling of productivity and compensation, however, things fell badly out of whack. One shaft kept spinning like a champ while the other started moving in slow-mo. If the economy were a machine, we’d send it to the shop.

Theory 2: We’re Not Measuring It Right

Another theory is that the pay-compensation gap is real but maybe not quite as large as the consumer-price-indexed compensation rates suggest. The BLS provides the following chart.

In this graph, the bottom dotted line is compensation adjusted using the consumer price index, but the light blue line above that is compensation that’s adjusted using something called the output price index, which is arguably more accurate. The authors of the article “Understanding the labor productivity and compensation gap” explain:

Workers are compensated based on the value of goods and services produced, not on what they consume. Using an output price deflator, a measure of changes in prices for producers, instead of the CPI is an alternative that better aligns what is produced to the compensation that workers receive. Each industry has its own unique output deflator that matches the goods and services that are produced in that industry.

By using these “deflators” for a variety of industries, they find that the size of the productivity-compensation gap “decreased in 87% of industries that previously showed productivity rising faster than compensation.”

To be clear, the gap isn’t going away if you use this technique, but it does typically shrink in most industries.

Theory 3: The Rich Got Most of the Pay Raise

The third and, to me, most convincing theory is that average folk had their productivity lunch eaten by their better off brethren.

This is clear when you look at the work by economists such as Erik Brynjolfsson and Andrew McAfee of MIT. In their book Race Against the Machine, they comment on a graph that shows the amazing and growing disparity between real median household income and real GDP per capita (which is one measure of productivity). Below is a more up-to-date version of the one they point to in their book:

They call it “striking” and then make this observation:

There have been trillions of dollars of wealth created in recent decades, but most of it went to a relatively small share of the population. In fact, economist Ed Wolff found that over 100% of all the wealth increase in America between 1983 and 2009 accrued to the top 20% of households.  The other four-fifths of the population saw a net decrease in wealth over nearly 30 years.

Ouch. So, yes, the productivity paychecks are real. And they do raise the standard of living — but not for everybody. Or even most people.

Were Gains by the Rich Earned or Stolen?

Of course, this raises another question: “Did those folks at the top earn that paycheck, or steal it?”

If that’s incendiary phrasing, don’t blame me. Blame the purveyors of conventional wisdom mentioned above. The implication has always been that we all benefit from productivity increases, but, in practice, as Brynjolfsson and McAfee say, “There is no economic law that says that everyone, or even most people, automatically benefit from technological progress.”

Maybe that makes sense? Let’s say a bunch of tycoon types invest in robotics to boost the productivity of the average worker on the line of some manufacturing plant. After the inevitable layoffs of many workers, do the rest of those surviving employees divvy up the compensation of the people who were laid off minus the cost of the machines? 

Probably not. Instead, the benefits accrue to the investors and the senior managers (especially CEOs) who made the decision to invest in the robots. That is, the rich get considerably richer while the surviving workers only get a modest increase. And the folks who were laid off? How much of a cut do you think they’re getting?

Yeah, bubkis. Or, in many cases, they actually lose economic ground.

Multiply this dynamic many times over the course of decades, and median incomes stay flat while GDP per person (which is an average rather than a median) goes up.

So, to answer our question, “They earned it, kind of, sort of, in a way, if you squint hard enough and quash any human instinct for justice and fairness.”

But at least we now have a clue about where benefits of the productivity increases go. That is the beginning of wisdom — and a fine antidote to fiscal fairy tales.

Productivity Chickens Coming Home to Roost

Recently, there has been a decline in U.S. productivity. In fact, some analysts claim that the U.S. has now seen five consecutive quarters of year-over-year declines.

The big question is why. There’s lots of finger-pointing. Some high-profile CEOs blame lazy work-at-home employees for the decline. Others argue, to the contrary, that it is the return-to-work policies that are most strongly linked to productivity declines.

There are plenty of other suspects as well. For example, many people switched jobs during the “great resignation” and so stepped into roles where they had to learn the ropes to become more productive again. Or there’s the rapid return of many employees back into the workforce, a dynamic often associated with temporary reductions of productivity.

There’s also the possibility that higher inflation — combined with pay increases that are insufficient to keep up with it — are simply demoralizing workers. Why should they worker harder for smaller paychecks?

And, of course, there’s the idea that younger generations just aren’t as eager as their older baby boomer counterparts to keep their proverbial noses to the grindstone. It’s less that they’re “lazy” and more that they just aren’t as willing to put up with bossism and toxic workplaces.

CEOs Venting Their Spleen

Meanwhile, CEOs have been venting their spleen about declining productivity, so much so that it feels as if there’s a new “leaked video of a CEO having a meltdown each week,” writes AJ Hess in Fast Company.

On one hand, I get their frustration. Their jobs are, of course, to boost the performance levels of their organizations.

On the other hand, what make these meltdowns both funny and sad is the extraordinary pay gaps between typical employees and their bosses. For example, recent figures indicate that S&P 500 CEOs averaged $18.3 million in compensation in 2021. That’s a whopping 324 times the median worker’s pay!

How did their pay get so exorbitant? Well, one answer is, of course, productivity. That is, they (and other upper-class Americans) have enjoyed the fruits of the productivity bumps of workers whose wages have largely stagnated over the last 40 years.

Which makes you wonder: If the typical worker had been receiving their full share of the benefits of productivity increases since the early 1980s, would we be in a position where “quiet quitting” was even a thing?

Maybe not. What we could be seeing is the productivity chickens come home to roost. If the rich get most of the monetary benefits of productivity increases, then let them do most of the work.

Or, at the very least, they — in partnership with the government — should stop whining and figure out a way to make productivity increases benefit everyone in their organizations, not just the investors and executives at the top.

100 Ways ChatGPT Could Help HR

How Can You Help, Chatbot?

I’ve been reading a variety of articles on the topic of how ChatGPT and similar AIs can help HR professionals do their jobs better. But I wanted to see what the AI thinks it can do. So, I asked about the ways ChatGPT could help HR professionals carry out their tasks in 20 different functions. It came up with the 100-item list you’ll see below.

To be clear, I asked the chatbot what it can do for HR, but I think what I ended up with is what it someday might be able to do.

Are You Sure You Can Do That?

When I read through the list the chatbot produced, I was skeptical that could actually do these things. So, to get a second opinion, I asked ChatGPT-3’s smarter younger brother Bing (based on ChatGPT-4) about it. Specifically, I asked Bing if ChatGPT could truly do the things it said it could do.

Here’s what it had to say:

Generative pre-trained models … have limitations, such as a lack of logical understanding, which limits their commercial functionality. According to some sources, generative pre-trained models can potentially assist HR professionals in some of the actions you listed…However, these applications are still experimental and not widely adopted in the industry. They also require careful supervision and evaluation to ensure the quality, accuracy, and reliability of the generated content. Moreover, they pose ethical and legal challenges such as privacy, security, bias, accountability, and transparency that need to be addressed before they can be deployed in real-world scenarios. Therefore, generative pre-trained models are not yet ready to fully replace human experts in HR compliance tasks, but they can potentially augment their capabilities and efficiency by providing assistance or support in some aspects of their work.

Let’s Put Some Numbers to That!

I thought that was a good if cautious summary, so I asked Bing if it could assign probabilities to the five compliance-related actions that ChatGPT said it could take on behalf of HR pros. Still sounding a note of caution, Bing stated, I cannot assign exact probabilities to those five actions, as they depend on various factors such as the quality and quantity of the data, the specific model and parameters used, the level of human supervision and feedback, and the evaluation criteria and metrics. However, based on my understanding of the current state-of-the-art in generative pre-trained models and their applications in HR compliance, I can provide some rough estimates of how likely they are to perform well on those actions today. These estimates are subjective and not based on empirical evidence, so please take them with a grain of salt.

Then, with those sober caveats properly laid out, Bing gave these estimates of how likely ChatGPT could really do these five things:

Researching and staying up-to-date with labor laws and regulations: High (80-90%)

Helping HR professionals ensure that company policies and procedures are compliant with relevant laws and regulations: Medium (50-60%)

Assisting in drafting compliance reports and audits, ensuring accuracy and completeness of data and information: Medium (50-60%)

Aiding in identifying potential areas of non-compliance by analyzing employee data and HR practices: Low (20-30%)

Providing guidance on best practices for compliance training and development programs: Low (20-30%)

These estimates are based on the assumption that generative pre-trained models are fine-tuned on relevant data and tasks, and that they have access to reliable sources of information. They also assume that the generated content is verified and validated by human experts before being used or shared. The estimates may change over time as generative pre-trained models improve or new challenges arise.

Being a nerd, I was tempted ask about the probabilities for all 100 actions, but since Bing correctly noted that the numbers were “subjective and not based on empirical evidence,” I thought that might be overkill. (Sidenote: I’m fascinated by the idea of subjectivity in an artificial intelligence).

100 Ways ChatGPT Might Help HR Pros

So, without further ado, here are the 100 things ChatGPT might, under the right circumstances, be able to do for Human Resources. I’ve listed them alphabetically by name of the HR function.

1) Compliance

  • Assist HR professionals in researching and staying up-to-date with labor laws and regulations.
  • Help HR professionals ensure that company policies and procedures are compliant with relevant laws and regulations.
  • Assist in drafting compliance reports and audits, ensuring accuracy and completeness of data and information.
  • Aid in identifying potential areas of non-compliance by analyzing employee data and HR practices.
  • Provide guidance on best practices for compliance training and development programs.

2) Compensation

  • Assist in analyzing market data to determine appropriate salary levels and benefits packages for employees.
  • Help in conducting salary surveys and researching industry compensation trends.
  • Aid in answering employees’ questions about their compensation and benefits packages.
  • Provide guidance on the design and implementation of employee benefits programs, such as health insurance, retirement plans, and paid time off.
  • Help HR professionals to create fair and equitable compensation policies and programs.

3) Diversity, Equity, and Inclusion

  • Assist in the creation of diversity, equity, and inclusion policies and programs.
  • Help HR professionals identify and eliminate bias in recruitment and selection processes.
  • Provide guidance on diversity and inclusion training and development programs for employees.
  • Assist in analyzing employee feedback and sentiment regarding diversity, equity, and inclusion.
  • Aid in identifying potential areas of bias or inequality within the organization.

4) Employee Benefits Administration

  • Assist in managing employee benefits enrollment, including answering questions and providing guidance to employees.
  • Help HR professionals analyze data to determine the effectiveness of employee benefits programs.
  • Aid in identifying potential areas for improvement in employee benefits programs.
  • Assist in managing employee benefits-related data, such as enrollment forms and plan documents.
  • Provide guidance on compliance with regulations related to employee benefits programs.

5) Employee Communications

  • Assist in developing and delivering internal communication strategies to employees.
  • Help HR professionals to quickly respond to employee inquiries and feedback.
  • Aid in analyzing employee feedback and sentiment regarding company news, policies, and initiatives.
  • Assist in managing employee communications-related data, such as email campaigns and messaging platforms.
  • Provide guidance on best practices for employee communications, such as creating clear and concise messaging.

6) Employee Engagement

  • Offering personalized recognition and rewards to employees
  • Providing guidance on employee wellness and work-life balance
  • Conducting virtual team building and social events
  • Providing information on opportunities for career growth and development
  • Generating reports on employee engagement metrics and trends

7) Employee Relations

  • Develop chatbots to help employees navigate the organization’s HR policies and procedures and answer common questions.
  • Use natural language processing to analyze employee feedback and sentiment to identify potential areas of concern and address them proactively.
  • Provide mediation services for employees involved in workplace conflicts through a virtual assistant.
  • Develop chatbots to handle employee complaints and grievances and ensure they are resolved fairly and consistently.
  • Use chatbots to provide employees with 24/7 access to HR support and guidance.

8) Employee Wellness

  • Use chatbots to provide employees with personalized recommendations on wellness resources and activities.
  • Develop wellness trackers that use natural language processing to track employee progress and make recommendations for improvement.
  • Use chatbots to provide employees with 24/7 access to mental health resources and support.
  • Provide personalized coaching and advice for employees on health and wellness topics based on their individual needs and preferences.
  • Develop personalized wellness plans for employees using natural language processing to analyze health data and identify areas for improvement.

9) Health and Safety

  • Use chatbots to provide employees with safety training and information on safety procedures and protocols.
  • Develop virtual safety simulations using natural language processing to simulate emergency situations and help employees prepare for potential safety hazards.
  • Use predictive analytics to identify safety risks and proactively address them before incidents occur.
  • Use chatbots to collect and analyze safety data and identify areas for improvement.
  • Develop chatbots to provide employees with real-time safety alerts and updates.

10) HR Analytics

  • Use natural language processing to analyze employee feedback and sentiment to identify potential areas of concern and address them proactively.
  • Create predictive models that use HR data to forecast employee turnover and identify potential areas for improvement.
  • Use chatbots to collect and analyze HR data and identify trends and patterns in employee behavior.
  • Develop chatbots to provide HR analytics dashboards that display key HR metrics and data in real-time.
  • Use chatbots to perform ad-hoc data analysis and answer HR-related questions quickly and efficiently.

11)  HR Information Systems (HRIS)

  • Provide real-time assistance to HR professionals in managing HR data, such as helping them locate specific employee information or assisting with data entry.
  • Help automate certain HR processes, such as scheduling interviews or generating reports.
  • Assist with employee self-service by providing instant answers to common questions, such as how to update personal information or check pay stubs.
  • Analyze HR data to identify patterns or trends, such as turnover rates or attendance patterns.
  • Help HR professionals stay up-to-date with the latest HR technology and trends.

12) Labor Relations

  • Provide legal guidance and assistance to HR professionals in managing employee relations and ensuring compliance with labor laws and regulations.
  • Assist with collective bargaining negotiations by providing data and analytics on industry standards and trends.
  • Help HR professionals navigate complex labor issues, such as union grievances or strikes.
  • Provide training and development resources to help HR professionals build their labor relations knowledge and skills.
  • Help HR professionals stay up-to-date with the latest labor laws and regulations.

13) Onboarding

  • Provide new employees with a virtual onboarding experience that includes interactive tutorials and videos on company culture, policies, and procedures.
  • Assist HR professionals in preparing and delivering onboarding materials to new hires.
  • Provide new employees with personalized assistance, such as answering questions about benefits or guiding them through the onboarding process.
  • Collect feedback from new employees on the onboarding experience to help improve the process.
  • Assist HR professionals in tracking and managing onboarding tasks, such as verifying employee documents or setting up equipment.

14) Organizational Development

  • Assist in facilitating the strategic planning process by generating ideas, identifying potential risks, and providing insights into industry trends and best practices.
  • Analyze employee data and HR practices to identify areas for process improvement.
  • Recommend process mapping tools and suggest process improvement initiatives that align with the company’s strategic objectives.
  • Provide recommendations for leadership development programs that align with the company’s strategic objectives.
  • Suggest leadership assessment tools, training materials, and coaching resources.

15) Performance Management

  • Assist HR professionals in setting and monitoring employee goals and objectives.
  • Provide real-time feedback to employees on their performance, such as alerting them to areas where they need improvement or recognizing their accomplishments.
  • Provide HR professionals with analytics on employee performance, such as identifying top performers or areas where the organization needs improvement.
  • Help HR professionals create and manage performance improvement plans for employees who need additional support.
  • Assist HR professionals in conducting performance evaluations and delivering feedback to employees.

16) Records Management

  • Assist HR professionals in managing and organizing employee records, such as resumes, job applications, and performance evaluations.
  • Help HR professionals comply with data privacy regulations by ensuring that employee records are stored securely and confidentially.
  • Assist HR professionals in locating specific employee records or information quickly and easily.
  • Provide analytics on employee records, such as identifying patterns in hiring or turnover rates.
  • Help HR professionals stay up-to-date with the latest record-keeping regulations and best practices.

17) Recruitment and Selection

  • Assist in creating job descriptions and job postings that are more engaging and attractive to potential candidates.
  • Help screen resumes and applications more efficiently, allowing HR professionals to focus on the most promising candidates.
  • Assist with scheduling interviews, sending reminders, and providing basic information to candidates.
  • Provide information to candidates about the company, its culture, and its values.
  • Assist in creating and administering pre-employment assessments, such as personality tests or skills tests.

18) Succession Planning

  • Help identify high-potential employees for key leadership positions based on performance data and other factors.
  • Assist in creating development plans and programs for high-potential employees.
  • Help identify skills gaps and training needs within the organization, and recommend training programs to fill those gaps.
  • Assist in identifying potential external candidates for key leadership positions.
  • Help HR professionals monitor and track the progress of high-potential employees in their development plans.

19) Talent Management

  • Help identify and source top talent from external sources, such as job boards or social media.
  • Assist in creating and administering pre-employment assessments, such as personality tests or skills tests.
  • Help HR professionals track employee performance and identify top performers.
  • Assist in creating career development plans for employees to help them grow within the organization.
  • Assist in creating and implementing employee retention strategies, such as recognition programs or career development opportunities.

20) Training and Development

  • Assist in creating and administering training programs, including creating training materials and quizzes.
  • Help identify skills gaps and training needs within the organization, and recommend training programs to fill those gaps.
  • Assist in creating and administering online learning courses and modules.
  • Provide personalized training recommendations based on employee performance and development goals.
  • Help HR professionals track employee progress and completion rates for training programs.

Final Thoughts

ChatGPT-3.5 is, let’s say, a very confident bot. But it’s not always an accurate one. Still, these robust large language models continue to dazzle and are doing things I never thought I’d live to see. So, while Bing and I are skeptical about 3.5’s current abilities, who knows where we’ll be in a year as HR technology vendors adapt these technologies to their own purposes?

My guess is that the AIs will continue to surprise us. Mostly in good ways, I hope.


A List of 22 Articles on ChatGPT and HR

Balancing the Pros and Cons of ChatGPT and AI in HR

Beyond Job Descriptions: 6 HR Tasks ChatGPT Can Do for You

ChatGPT – A tool for accelerating innovation, but not a panacea

ChatGPT 101 for HR Pros

ChatGPT and the Future of Human Resources: A Step-by-Step Guide

ChatGPT can improve HR functions, but not without risk

ChatGPT in Human Resources

ChatGPT Takes Over HR

ChatGPT: ‘Bigger than anything’ HR has ever seen?

ChatGPT: Is this the next disruptive trend in the HR space?

ChatGPT: The Secret To Helping HR Save Time And Reduce Stress

ChatGPT’s Applicability Toward the HR Department

How ChatGPT Can Be a Game Changer in Human Resource Management

How useful is ChatGPT-style AI for HR?

How will ChatGPT impact HR?

Leveraging ChatGPT in HR Activities: How Automation Can Support HR Upskilling

One-half of HR leaders evaluating ChatGPT guidelines

Performance review cycles are tough—could ChatGPT help?

The 7 Best Examples Of How ChatGPT Can Be Used In Human Resources (HR):

Third of HR professionals want to use ChatGPT at work, exclusive data reveals

Uses Of ChatGPT In HR

What ChatGPT and other generative AI tools mean for HR