Walmart Is Changing Its Labor Model: How Many Workers Will Lose Their Jobs?
作者：Tim Worstall @ 2015-9-06
Walmart is quite significantly changing its labor model. Moving from a near hire any live body and let them get on with it one to something where people are well trained, well paid and presumably of rather higher productivity.
This is what many have been crying out for the company to do for years of course: move to something closer to the Costco model than the one that Walmart has traditionally pursued.
However, as some like me have been pointing out all along there is a flip side to that change in models. Which is that the end aim is of course to employ fewer of those more productive people at those higher wages.
The point being that if one can raise productivity levels by more than the increase in cost then of course profits will rise: that being the end goal for all shareholder owned companies.
The interesting question is going to be how many people either lose or don’t get jobs as a result of this strategic change: and I’ll suggest a method by which we can calculate this a little later.
There’s two prongs to Walmart’s new approach. The first is to take a leaf from Henry Ford’s book with that $5 a day thing
One motive is better public relations at a time when inequality is a hot-button political issue. But bottom-line calculations also play a role.
Employee turnover costs money—by industry estimates as much as $5,000 per front-line worker, or 20% to 30% of an entry-level salary.
Standard turnover in retail is 50% in the first six months. If Wal-Mart can reduce this churn, persuading people to stay at least 12 to 18 months, it will save “tens of millions of dollars a year,” according to Ms. Oliver.
零售业员工在工作头六个月内的流动率通常是50%。如果沃尔玛能减少这种流动，劝说员工至少工作12到18个月，照Ms. Oliver 的估计，“每年能省下数千万美元”。
As I explained way back here, Henry Ford’s $5 a day
was not what most people think it was. It was most certainly not, as all too many would have it, so that the newly richer workers could all buy a Model T.
That would have been a great way to lose lots of money. A company cannot pay its own workers more, then see profits rise as they spend that cash on the company’s products.
This is trying to raise yourself by your bootlaces. It also wasn’t about trying to create a vibrant midle class. What it was about was reducing the job churn on the assembly line.
Ford was getting through 50,000 workers a year in order to have 13,000 working on the line at any one time. That had vast recruitment and training costs.
So, that’s what Walmart is doing here. Let’s see if we can reduce those costs by having less churn.
That, in turn, means perhaps not bottom fishing in the labor market but improving pay relative to others so that people will stick around a little longer.
This could well be a good move too but only time will tell.
There’s also a second prong to the new strategy:
Front-line employees—cashiers, cart pushers and sales associates—will now spend their first months at the company in a supervised on-the-job training program.
In the past, they sat through a few days of orientation and safety drills, many of them focused on compliance with environmental and health regulations.
The only real job training happened in the store—knowledge passed on by more experienced employees.
There’s two sides to this. One is the obvious point that if you’re expecting your workers to stick around longer then you’re also going to be willing to invest in them rather more.
Because you’ll be able to amortise your investment in them over that longer period that they’re working for you.
And there’s the more obvious point of that end goal: better trained workers will be, ceteris paribus
, more productive. And thus we can see that Walmart is trying to move from one labor model to another:
Economists who study retail distinguish between “low-road” and “high-road” employers. One group keeps labor costs down, the other invests more in workers and reaps the benefits in higher productivity. Cost-conscious Wal-Mart is trying to move toward the high road.
This is all entirely traditional labor economics by the way, there’s nothing mysterious about any of it. However, there is a sting in the tail here.
For well over a decade now I’ve been pointing out that yes, sure, Costco pays its workers very much better than Walmart does. But it also uses, per unit of sales, about half the labor that Walmart does.
Thus the shouting that Walmart can and should pay its workers like Costco does comes with that sting in that tail: for moving to the same pay structure would entail at least attempting to move to the same productivity levels.
Meaning that Walmart would employ about half the number of people per unit of sales than it currently does.
And now we’re seeing that Walmart is taking at least baby steps to that higher road labor model. And the interesting thing is going to be, well, is the prediction about employment levels going to come true too?
Just in a little more detail. Productivity is the amount of work (really, the amount of value added) that we get from one hour of labor. Raising productivity thus means getting more value added from one hour of labor.
And if sales are static that then obviously also means using less labor per unit of sales.
Thus raising productivity is the very same thing as saying that less labor is going to be used. This still holds even if sales or output rise: there’s still less labor going to be used than there would have been at the earlier, lower, level of labor productivity.
And the way to test it is pretty simple, because we can find the numbers we need to measure labor productivity in the Walmart accounts.
We know the number employed in the US….some 1.4 million….and we know what sales are in the US…$288 billion….so labor productivity is $205,000 and change per worker.
That’s actually sales not value added but that still gives us what we want, a number to compare over time (Costco’s sales per employee are about double this).
As labor productivity rises as a result of more training and lower churn from the pay rises then we would expect to see this number rise.
More sales per employee. And then we will also be able to calculate how many jobs have been lost to this rise in productivity.
For, say, that sales rise to $250k per employee. We can then calculate how many employees would have been needed if productivity was still the old, lower, number.
The number of jobs lost will therefore be the difference between the number actually employed and the number who would have been without the productivity gain.
Yes, obviously, we would need to discount this for the general inflation rate.
My prediction is that productivity will indeed rise at Walmart in the coming years. And also that sales per employee will rise, meaning that the number employed will fall.
Not fall necessarily from the current absolute level, but fall relative to where it would have been absent the productivity increase.
Anyone want to bet against that prediction?