The Magic Beans Problem.

How do you build workforce programs for employers who want to have fewer jobs?

The issue.

The people driving the ship for large American employers are betting that AI will help them make more money by killing as many jobs as possible. How do you build workforce programs for that?

Explain.

Here’s an unfair illustration of an email I feel like I get once a week:

Dear Nick:

Jobs That Work Solutions is killing it in the employment consulting game. Your company is making waves for how it advises companies, foundations, nonprofits, and governments on how companies, foundations, nonprofits, and governments can obtain jobs in government. It’s really impressive and I didn’t just base this on a 30-second review of your LinkedIn.

We know how hard it can be to scale your business while dealing with the constant pain of paying people money for their work. Here at Bēnstalk, we help growing companies like yours offload the unnecessary time and work of having a staff by using our proprietary AI to do AI things that mean you don’t need a staff anymore. In fact, our AI has been shown to increase sales and productivity and handsomeness by 957%.

When can we schedule a demo so that you can make all the money forever?

Cheers,

[Double Check that Sales Rep’s Name Populated Before Sending]
Sales Representative
Bēnstalk
”The future of tech is the future.”

___________________________________

From: Boss@beanstalk.tv
To: [Double Check that Sales Rep’s Email Populated Before Sending]@beanstalk.tv
Subject: Re: Nick

For this real email, I just finished reviewing the file on Nick. Go ahead and send him a message right away so he can use our AI to make all the money forever and go to a place where everything is beautiful and nothing hurts!

The Boss
Boss
Bēnstalk
”Success is stumbling from failure to failure with no loss of enthusiasm.” —Gary Oldman

I have started calling these “Magic Beans” emails. When you read or hear these pitches, you can mentally replace “AI” with “magic beans” or “horses” or “Arby’s Beef ‘N Cheddars” and you’ll pretty much get the same amount of detail and logic as to how the pitcher’s AI lets you fire everyone and do the best business that’s ever been business’d.

Big business and its investors are hearing this pitch about AI a lot, too. While they may not be buying precisely what’s in the email above, they’re definitely buying into the Magic Beans being able to kill a bunch of jobs they don’t want to pay for—even if they don’t exactly know the fine details of how all that is supposed to work.

They maybe kinda might’ve staked the economy on it. From The New York Times:

The stock market run-up — the S&P 500 is still up about 14 percent this year despite the recent shivers — could foreshadow widespread economic gains. But [Kenneth Rogoff, a Harvard economics professor,] doesn’t think that is the case.

“A big part of the high stock prices is not a reflection of high future growth,” he said. Rather, it is a sign that A.I. is expected to boost productivity and shrink employment. “The firms all think they’re going to shed a lot of labor, and that’s why the profits will be high,” he said.

One, Firing Everyone would be bad. Seventy percent of the American economy is based on consumption. Strangely, people who don’t have jobs—and therefore no dependable income streams—have a “very subtle loss of interest in shopping,” as one financial services CEO described the straits of feds not getting paid during the government shutdown.

Two, investor pressure to embrace The Magic Beans That Let You Fire Everybody hacks at a fundamental assumption at the heart of workforce development.¹ Workforce development is about getting people who need jobs into jobs that need filling. Big employers—ones with plenty of sway in the current political conversation—continue to assert that they have jobs they haven’t filled because they can’t find skilled talent and they need public funding to build the programs that help them make that talent.

But is it a good idea to invest in big companies claimed “skills gaps” if the current environment places so much pressure on them to also get rid of a mess of jobs, too? And if so, how can we make sure our investment is worth it?

Falling off the beanstalk.

MIT’s Zeynep Ton kicked off a lot of the current good jobs movement through her 2014 book on how companies could make more money investing in their workers. Based on the success of companies like Costco and QT, a key part of Ton’s pitch was that when a company takes care of its workers, the workers better serve customers—the people who are the source of the company’s revenue. Happy customers tend to attract more customers, which results in more revenue.

In 2023, she published a follow-up that talked about her experiences counseling employers in the years since publication of her first book. She writes about how her pitch just didn’t land with several executives. One told Ton that her company didn’t see the people and companies who consumed their products as the main way the company made money—so of course they weren’t interested in investing in workers who had the most customer contact.²

[The company] had focused on creating scale by acquiring other companies. They spent extravagantly on advertising their services. Running the core business well and improving the quality of service to customers was not a priority.

That’s a telling passage for any number of things in the American economy. Yet, the one that sticks out to me is how the way a company thinks it makes money flavors how it hires and treats workers—and what that means to their talent development if the expectation is that AI magic beans will turn publicly traded companies into big multinational profit dispensers at the expense of workers.

If you take a browse at the business press, you’re likely to read about a corporate CEO or four saying they have amazing, well-paying jobs that no one is willing or able to take. On the same site, you’re also likely to read the same corporate CEOs talking about how AI is going to kill all the jobs. Both things are key drivers in how political leaders are thinking about jobs policy at the moment.

Yes, there are some key distinctions in those two assertions: the jobs usually raised in today’s skills gap conversation are ones requiring physical labor to make things, while office and operational roles are the ones executives think The Robots can do more cheaply and effectively than us humans. But the same company leaders, probably under the same pressures to push down overhead, are making decisions about both issues. I don’t think you can divorce the process of making calls on one from another.

For example, one way that employers facing skills gaps could find more talent is by investing in things like skills-first hiring, which requires HR systems and evaluation strategies that identify and assess the skills that applicants have. That’s overhead. Investors pressuring companies to cut jobs aren’t likely to differentiate between the overhead the company needs and the overhead it can reasonably shed. They just see overhead.

That’s painful from a workforce development perspective. But if the money expects an employer to have fewer jobs, why would they spend to fix the problems that need to be fixed to fill the jobs they have open?

So what should we do about it?

Well, for one, I’m not as pessimistic as all this might read. I think there is a way forward, but it involves asking more questions than we ask of employers right now when workforce programs’ time and money.

Something fascinating this stirred up from my memory that’s important context: in Trump I, there were political appointees were pretty vocal (and public) about their skepticism about government subsidizing training programs for specific employers. They thought that these companies should have started these programs years ago, given how dire the companies professed their talent needs were. The appointees didn’t like the idea of fronting costs that the appointees thought the companies would recoup and then some if they just spent the money.

Interestingly, the type of investments Trump I questioned are pretty much the signature workforce spending strategy of Trump II. Since July, the Administration has invested $122 million in pay-to-train awards meant to incentivize employers to start training programs or retain staff so they don’t lose their jobs to things like AI. There are hooks. For one batch of money, grantees may only pay back employers what they invested so long as they don’t Fire Everybody for six months.

But given the pressures employers are facing based on the smallest hint AI could greatly reduce their overhead, I do wonder if that’s enough to make the investment worth it. I don’t agree with the broad brush approach of those Trump I folks, but I do think answering questions layered in their argument has value in the current environment.

If I was spending public money in this current environment, I would want to know more to be sure that the employer was a partner worth the risk—and capable of returning the government’s investment in the way of continuing to hire people. I also would want to know what the employer has done to try to fix their current talent gaps, and I would want a good indicator that there was a non-government funding stream that could sustain the program when the public dollars go away.

And if I was spending toward employers retaining workers under their company’s AI strategy, I would want to know what is that strategy. How will leaders make decisions on who has to go? Are their decisions to get rid of workers are logical and based on an understanding of what AI is—not a third of an idea of what it could become?

Because you know why there are so many questionable Magic Beans emails like the one above?

Sometimes they work.

Card subject to change.

I hope you had a Happy Thanksgiving. My gift to you this holiday season is a back-to-work newsletter about a potential economic apocalypse. You’re welcome.

For Friday, I’ll be writing about new Trump Administration guidance on workforce funding that sure seems to say a lot about where the money is heading. Congress is also out there lurking, ready to do absurd things to workforce money at any time. I’ll try to give you a sense of where all that’s at.

1. You might be asking, “Aren’t businesses always trying to spend the least amount on overhead?” As a small business owner, I can confirm that premise. But betting on job cuts that aren’t really rooted in reality yet, but investors are likely to aggressively expect, is something else. And something worth being concerned about given how much the idea of AI is keeping the economy going and real questions about whether AI-boosted companies’ value is artificially inflated.

2. There’s a whole ocean I want to sail on this issue, but something that Ton also talks about is the very low opinion executives had about the value of their workers. That sure seems to speak to why business leaders are so sure that they can replace talent with whatever AI—and why many early efforts to do so haven’t worked.