Will the world really be 50 million workers short by 2030?

This BBC podcast is supported by ads outside the UK.

Before Prodigy, I tried everything to get my Gen Alpha students into math.

I even tried to talk like them.

No cap, class.

Multiplication is low-key, a cheat code for math that'll give you sus levels of aura.

Bet.

Yikes.

But now with Prodigy, I've got the number one math motivator on my side.

As they play and up their skills, I get to follow along and gain real-time insights with my free teacher dashboard.

They even call me the Rizzler now.

I think that's good.

Learn more at prodigy.com.

Hello and thanks for downloading the More Orless podcast with a program that looks at the numbers in the news, in life, and in your jobs.

I'm Tim Harford.

Loyal listener Robert Gibbs emailed in to ask about a stat he heard in a talk by Jensen Huang, the CEO of NVIDIA.

NVIDIA is a US tech firm that's already become the most valuable company on earth by selling advanced chips to the companies developing generative artificial intelligence.

Huang, its boss, was speaking at a conference in California where he told the audience, We know very clearly that the world has severe shortage of human laborers, human workers.

By the end of this decade, the world is going to be at least 50 million workers short.

Huang was making the case for the futuristic solution to this problem, humanoid robots.

But is the claim correct?

Is the world really going to be 50 million workers short by 2030?

Eager to justify our own jobs in journalism, we asked Nvidia where Huang got this figure from.

They didn't reply.

Possibly due to vacancies in their public relations department.

But we have tracked down a figure that looks remarkably similar, in news articles from India citing a report by Boston Consulting Group, also known as BCG, a massive global management consulting firm.

The report hasn't been made public, but its author, Rajiv Gupta, agreed to talk to us.

So, did he recognise the figures the NVIDIA boss has been using?

We do not know if he's using our report, but yes, the number he uses does match the number that we have in our report.

Okay, let's get into it.

How did Rajiv work out that there would be a shortage of 50 million workers by 2030?

Like most estimations, it is a gap between what we project as the demand for jobs and the supply for jobs.

Simple.

First up, estimate the demand for jobs.

On the demand side, what we have done is we have taken the current employment data accurate as of 2024 plus all the open job postings.

Using data from sites such as LinkedIn, Rajiv and his team worked out how many unfilled job vacancies there were in the whole world as of 2024.

He reckons there are currently 10 million.

Now this figure is obviously a little ballparky already as not all jobs are advertised this way.

We then forecasted the GDP growth of different countries, adjusted it for labor productivity and adjusted it for any impact of AI and technology to come out with an estimated employment growth rate.

Not so simple.

These extrapolations hinge on predictions about the future of AI, but no one really knows what's going to happen, with some tech soothsayers predicting unimaginable transformations over the next five years, and some others suggesting it won't make much difference.

Reggie's analysis takes a middle way.

He decided to treat AI like any other tech disruption, like the internet, for example.

And he went through different types of jobs, projecting the likely disruption for each area.

And in some kind of industry across skills, the impact is as high as 50%.

In some, it is as low as 2-3%.

So, based on which industry, what type of work we have estimated, what kind of impact that can have.

Some types of work are going to be hit hard by AI, such as software development and testing.

Some, not so much, such as construction.

And there'll be some new types of jobs too, for example, AI prompt engineers.

So, overall.

We have a point of view that the worker skills will change dramatically because of AI, but net it's going to create many more new jobs than it's necessarily going to take away.

Okay, back to the 50 million.

That's the first step done, a necessarily massively simplified analysis of five years of economic growth and disruption that gives you one plausible projection of the likely employment demand in 2030.

Then you have to work out how many workers there'll actually be.

We took the labor flow rate data, so different countries have different labor flow for women, for people of different ages and so on, and we adjusted it for both retirement and new entrepreneurs to calculate the supply side of labor based on skills.

They looked at the demographics of all the countries in the world and the likely flow of new workers entering the system and older workers leaving it.

Some countries have aging populations and declining workforce trends.

Some don't.

They worked out three scenarios, conservative to aggressive, and came out with three estimates.

The numbers were close to 30 million, 40 million, 50 million.

But what does Rajiv actually mean by a shortage of workers?

Is he predicting an end to unemployment, that everybody will have a job?

and the world will still need 30 to 50 million more workers?

Well, no.

We not going down and looking at what will it do to the unemployment rate, we're just looking at the skills for open roles that have not been filled.

So there could indeed be very high unemployment in some places and in some countries alongside a worker shortage.

And this is where AI comes in again, because the worker shortfall is biggest in areas where AI is going to find it hardest to reach.

60% of the shortages comes from what we call blue-collar industrial workers.

So manufacturing, mining, cleaning, private security, then in sales professionals, retail, store managers, and so on.

Then healthcare and education was another 12 to 15%.

Office administration was 10%

and white collar workers, technology and so on was another 10%.

And why are these sectors going to be short of workers?

It is the preference of workers in different countries and what they want to.

So while we talk about labor shortage, there are areas, white collar, highly skilled being one, where there is likely to be a surplus of labor as well.

But it is driven by the growth in some of these industries, let's say construction, manufacturing and so on, not being filled by individuals who necessarily want to get into that area.

And to be clear, the worker shortfall isn't everywhere on the planet.

It's concentrated in specific rich economies.

So 90% of the labor shortfall comes from 20 countries out of the 200 plus.

Obviously, just given the size of the economy, USA is one large such country.

After US, UK,

Germany, South Korea, Russia are the next, and Poland, Hungary, Czech Republic put together.

These are the next four or five countries slash regions.

And then followed by Japan, Australia, Canada.

The bottom line is that Rajiv obviously doesn't have a crystal ball that's telling him how the global economy is going to shift and change over the coming years.

But he was able to roughly trace out what happens if you take the current trends and spool them forward a few years under a set of assumptions.

See, most estimates and extrapolations do turn out to be wrong, and I'd be humble enough to accept that.

However, what this provides is a framework to think in what industry in which country, what skills are likely to have a gap, so that organizations and countries and workers can start thinking about it and maybe geared up towards it.

One part of that might be that people are paid more to do the jobs not enough people want to do.

Although NVIDIA boss Jensen Huang thinks a pay rise won't solve the problem, and so we're going to need robots to do those jobs.

But there are other options too.

Higher levels of immigration could be used to fill these jobs, but that comes with all sorts of practical and political challenges.

A forecast of 45 to 50 million cap in labor is assuming no significant interventions.

But actually, the report then goes on to recommend interventions.

So, I do hope and pray that by 2030, the number is proven wrong and those interventions have actually happened.

Our thanks to Rajiv Gupta from Boston Consulting Group.

And that's all we have time for this week.

But if you see any suspicious stats, please let us know at more or less at bbc.co.uk.

We will be back next week and until then, goodbye.

This episode is brought to you by Allianz Travel Insurance.

You're so excited to finally make it to Ireland.

But the stomach bug you caught in Dublin wasn't exactly on your bucket list.

Benefits like emergency medical protection can have your back.

Learn more at AllianzTravel Insurance.com.

In the market to sell or trade in your luxury exotic, call the company the dealers trust.

Gimme the vin.com.

Looking to sell your Hurricane, 9-11, Cullin Inn, Ferrari, or Bugatti.

GimmeTheVin will pay you the most.

It's as simple as going to GimmeTheVin.com and entering your car's VIN number or license plate number.

You'll get an offer on your car in minutes, well above any dealer quote.

You can go to one of their offices or they'll come to you.

GimmeTheVin.com is America's best car buyer.