Five years after the worst global financial crisis since the Great Depression, we are now constantly being assured that a manufacturing revival is occurring in the United States at the moment. Or about to. Not so much in Europe, but Europe appears to have its own issues these days—as does Asia, with growth in China slowing to a pace that would be phenomenal if it were to occur in the US or Europe, but in fact isn’t fast enough for China to maintain current employment levels. So all eyes to the US.
There is, in fact, some evidence for an improvement in US manufacturing. But it’s not as compelling as supporters would imply, and it’s not evenly spread across all sectors of the economy. In fact, it’s pretty weak. For example, the growth in manufacturing-related jobs remains abysmal, even as manufacturing output picks up. In addition, much of the optimism for job growth on the back of cheaper energy has yet to be validated. In fact, the bright folks over at FT Alphaville have opined that the entire shale gas = more jobs story is pretty much a myth. More to the point, they are dubious about claims that the shale gas “revolution” has sparked the US manufacturing “revival” in the first place. As the NY Times notes in a separate article, profits are rising, not job growth.
More jarringly, job growth in general seems to be recovering faster in other countries than it does in the US. And it appears to be to just some, but quite a few countries:
While several European countries have fared worse, Canada, Sweden and even Britain, which is trapped in yet another recession, have enjoyed healthier job gains than the United States. In fact, of the nine countries surveyed by the Bureau of Labor Statistics, only perennially-troubled Italy and Japan performed worse.
The Times points out that, for example, “Although the construction field gained 69,000 jobs in the first five months of 2013, with 5.8 million jobs in May, that was still nearly two million fewer jobs than in 2007, according to the Labor Department.” Ah, we’re getting somewhere here—the construction industry. Remember the 2000s? When homebuilding seemed to be driving America’s economy? That’s because it was a much higher percentage of US GDP during the 2000s—around 6%—than previous periods. Currently, it’s less than 3%.
But it’s not just the construction industry that explains what’s happening. The Times article notes further that other countries have generated jobs on the basis of strong exports.
Germany’s economy, for example, was powered until recently by shipments of machinery, cars and other products of its high-end manufacturing industries. Australia emerged largely unscathed from the downturn, thanks to booming Chinese demand for raw materials.
The German government also went to great lengths to discourage outright layoffs, instead encouraging employers to keep workers in a part-time capacity. At the same time, letting workers go in Europe is a much more costly proposition for big employers than it is in the United States.
And there’s another wrinkle. US industrial production has grown twice as fast as US GDP since the start of the recovery. And this while the industry has been adding jobs at an anemic pace, if the numbers tell us anything. But it’s mostly explainable by the fact that from 2007-2009, manufacturing output dropped by almost 15%, adjusted for inflation. So the 16% growth in manufacturing output since 2009 has mostly recovering been what was lost, with a little bit extra.
And then there’s the really troubling issue:
Another sign of distress is the persistence of long-term unemployment four years into the recovery. While the number of workers who have been out of a job for less than five weeks is almost unchanged from 2007 levels, there are roughly 4.4 million Americans who have been unemployed for more than six months, a 257 percent increase since 2007.
So, where are these people going to find jobs? And what kind of jobs will they find? In particular, will they find jobs that are comparable in pay to the jobs they lost? Well, not if robots keep taking the new jobs. Robots? Really?
Well, maybe, but I suspect that it’s probably too soon to tell. But it may not be too soon to start thinking about this, and maybe worrying a bit as well. Manufacturing remains an important (although still declining) share of the US economy—just as it does in a number of European countries. Except the rate of decline in European countries has been much, much slower than in the US—with the UK being possibly the major exception here. But manufacturing in particular is important for a number of reasons, which are well-understood by economists. Historically, manufacturing jobs have been the way into the US middle class. It was what provided fodder for the American dream. And it still has significant multiplier effects on the rest of the economy—that is, every $1 generated by manufacturing also generates a certain amount of additional income elsewhere in the economy, more so than most other industries.
Some recent reports from management consultants, who pay attention to this sort of thing, highlight the continued importance of the manufacturing sector, as well as what’s been happening to it over the past several decades. The first, from McKinsey and published in November 2012 (Manufacturing the future: The next era of global growth and innovation), points out that developing economies, many of which are in the process of developing consumers in substantial numbers, will continue to drive global demand for manufactured goods. McKinsey opines, in fact, that over the next 15 years an additional 1.8 billion people “will enter the global consuming class.” And they’re all going to be buying stuff—and someone is going to be making the stuff that these consumers will want to buy. But manufacturing is no longer (if it has ever been) a homogeneous entity. In fact, McKinsey has an interesting categorization of manufacturing companies, one which intuitively makes sense, and which explains much of what has been happening over the past several decades—and allows for some interesting thinking about what may happen next.
McKinsey first points out what many of us who following manufacturing industries, including public policy makers, have understood for years—there’s a definite curve here:
Globally, manufacturing continues to grow. It now accounts for approximately 16 percent of global GDP and 14 percent of employment. But the manufacturing sector’s relative size in an economy varies with its stage of development. We find that when economies industrialize, manufacturing employment and output both rise rapidly, but once manufacturing’s share of GDP peaks—at 20 to 35 percent of GDP—it falls in an inverted U pattern, along with its share of employment. The reason is that as wages rise, consumers have more money to spend on services, and that sector’s growth accelerates, making it more important than manufacturing as a source of growth and employment.
The sector is also evolving in ways that make the traditional view—that manufacturing and services are completely separate and fundamentally different sectors—outdated. Service inputs (everything from logistics to advertising) make up an increasing amount of manufacturing activity. In the United States, every dollar of manufacturing output requires 19 cents of services. And in some manufacturing industries, more than half of all employees work in service roles, such as R&D engineers and office-support staff.
As advanced economies recover from the Great Recession, hiring in manufacturing may accelerate, and some nations may even raise net exports. Manufacturers will continue to hire workers, both in production and nonproduction roles (such as design and after-sales service). But in the long run, manufacturing’s share of employment will remain under pressure as a result of ongoing productivity improvements, faster growth in services, and the force of global competition, which pushes advanced economies to specialize in activities requiring more skill.
Those jobs requiring more skill? This is where the McKinsey categorization comes in handy—where will those be? Here’s another useful comment:
No two manufacturing industries are exactly alike; some are more labor- or more knowledge-intensive. Some rely heavily on transportation, while for others, proximity to customers is the critical issue. We have identified five broad manufacturing segments and analyzed how different production factors influence where they build factories, carry out R&D, and go to market.
The largest segment by output (gross value added) includes industries such as autos, chemicals, and pharmaceuticals. These industries depend heavily on global innovation for local markets—they are highly R&D intensive—and also require close proximity to markets. The second-largest segment is regional processing, which includes industries such as printing and food and beverages. The smallest segment, with just 7 percent of global manufacturing value-added, produces labor-intensive tradables.
And here’s the chart that makes the point even more forcefully:
McKinsey, I have to say, puts out some very good research on economic and political issues, which is what the future of manufacturing entails—economics and politics. A more recent report from Deloitte, in conjunction with the World Economic Forum and published just this past month (Manufacturing for Growth: Strategies for Driving Growth and Employment), is equally optimistic about the future of manufacturing, but it also just points up how good the thinking at McKinsey is on this issue. Still, it’s good to know that people and institutions continue to pay attention.
Because even those of us who think that austerity remains a bad idea are struggling a bit to find a place where increased spending on “jobs” should actually go. I mean, I’m sympathetic to Paul Krugman when he says that jobs are critical now, and not the deficit. I just don’t understand what jobs he’s talking about—and to be fair, he understands this problem as well. Infrastructure jobs take a long time to develop—yes, there’s a critical need for fixing bridges in the US, to take an example. Or to expand existing rail services, especially mass transit systems. But these all take years to get going, even if Congress were in a mood to support such measures, which this Congress clearly is not. And if one were to push for “jobs” as a political program, as appealing as that may sound, I have a real difficulty seeing where credible policy options are. Think back on that McKinsey chart—hiring people to do labor intensive jobs that have no obvious career path gets them off the unemployment rolls for a while, I admit, but then what? To paraphrase Michael Dukakis’s insight in his otherwise disastrous presidential campaign, we can’t all be knowledge workers for each other, any more than we can all be making hamburgers for each other.
But there’s a more fundamental transition going on as well, I think. The McKinsey folks are deeply aware of this, but not quite sure what to do about it. When I was coming of age back as a teenager back in the 1960s (cue laughtrack), jobs were pretty plentiful. And that was jobs for teenagers. I was fortunate in that my father had a small construction company, so I spent my summers and occasional weekends doing construction work, sometimes with a union permit. But in general there were lots of jobs. Pumping gas at service stations was a pretty good one, especially if you wanted to become a car mechanic. Working as a cashier at the local supermarkets was another. I can still recall bumping into high school classmates doing both of these things.
Times have changed. Both of those job categories now are considerably less plentiful. In most places you need to pump your own gas now, like it or not, and supermarket checkout scanners are everywhere. There are still cashiers in supermarkets, but not nearly as many as there were back in the day, and I don’t imagine the summer job prospects are as good as they were then. And if you want to become a car mechanic now, you probably need to get your degree in electronics before you even think of opening the hood on anything coming out of the car companies these days. And that’s not all, of course—ATMs have largely replaced bank tellers. And when is the last time you dealt with a travel agent? Every time I wander into some branch of my bank and do a transaction at the counter, someone always points out that I could have done the same transaction through the machine. At which point I mention that the more I use the machines, the less need the bank will have for them. Conversation usually stops at that point.
Of course, it’s not clear to me that supermarket cashier jobs actually led to anything. I’m not sure that much of anything that happens in a supermarket is a particularly high-skill endeavor. But hanging out in with the car mechanics? Well, yes, that could generate work and a career—you start picking stuff up just by listening, then you’re changing carburators, and eventually you’ve got enough of a knowledge base to apprentice yourself somewhere, and get real work. And this was back when real work actually meant doing something with your hands, and could still lead to a life that embodied the American Dream. And not just car mechanics. American life—indeed, life in many places—used to be full of similar opportunities for this sort of apprenticeship. Not any more. Unless we’re talking about home building—and we saw what happened there. We’ve discussed the issue of what’s real work before, but I don’t expect we’re any further along on the topic than we were a couple of years ago. But technology continues to undermine capital, as Isabella Kamisnky, one of the bright lights over at FT Alphaville, has been reminding us.
This is one area where America and much of Europe has parted ways over the past several decades. While America has been busy trying to put its unions out of business, Europe has given them a place at the table. Companies in Germany and Sweden, where manufacturing continues to thrive, have union representatives on the Boards of Directors of manufacturing companies. And apprenticeship programs are alive and well. Britain, in fact, is now looking to re-establish apprenticeship programs in areas where the skill-set for something is in danger of disappearing in a generation—how to repair a thatched roof, for example, or rebuild fireplaces of certain designs. You can’t reverse engineer everything, it turns out. In Britain, these traditional crafts apprentice programs aren’t government programs—in this instance, it’s the National Trust, which maintains properties of national interest. But the British government has its own set of apprenticeship programs it’s expanding, even in the face of its not-very-successful austerity programs.
Which brings us to a further issue—robots. This is an area that has already had an impact on productivity and jobs growth, so much so that the very smart folks over at FT Alphaville have been devoting a whole series of posts to the issue. Here’s the deal in a nutshell—robots are getting better at performing what humans can do. There’s quite a lot of handwaving here, with a number of robotics developers indicating that the robots they’re developing aren’t meant to replace humans in the manufacturing (or whatever) sector. Still, as Mandy Rice Davies put it so eloquently in another context, they would say that, wouldn’t they? But the developments have been very impressive from a technological point of view.
And before we get too Luddite about all of this, let’s concede that there are any number of scenarios where having robots do something would be preferable to having people do them simply because the activity in question is so dangerous. Cleaning out abandoned and collapsed mineshafts comes to mind. Picking up slag in an aluminum smelter. Entering a burning building where a child is thought to be hiding. Open pit gold mines in Africa and South America, where safety regulations, if they exist in the first place, are barely enforced. Let’s face it—there are some jobs where it would be a better idea for robots to do them than for people to do them. In fact, we’re starting to see robots in agriculture, a market where it’s getting more and more difficult to find workers, especially for the more back-breaking jobs.
Indeed. As the FT Alphaville series shows, robotics are becoming proficient at any number of things. Still, the technology sector in general continues to assure everyone that, as a recent note in Technology Review states, “technology will actually boost the U.S. economy and create more jobs, even if some jobs do disappear forever.” And, in fact, we keep hearing this—Technology Review itself had a breathless piece last year extolling the virtues of a new category of robots, and the potentially positive impact these could have on US manufacturing.
This of course sounds vaguely familiar—businesses and the robotics/automation industry have been telling us this for some time. And it’s not as if there isn’t some supporting data, although it’s complicated:
The International Federation of Robotics, a trade group, likes to highlight a study from 2011, which found that in Brazil, South Korea, Germany, China, and the U.S., employment rates rose even as industrial robot use grew. However, the robots being counted in that study were dangerous, dumb versions that can be used for a very limited range of tasks. They can’t work alongside or with people. The robots being brought forward by Rethink Robotics and others are headed to a much broader range of workplaces may have wider-reaching economic and social effects.
Well, if you introduce a new class of smarter, more adaptive robots into the manufacturing process, and, say, couple that with 3D printing, in a domain that has already been significantly transformed by the introduction of computer-aided process control, you’ve created the potential for a significant amount of economic benefit for some (rentier heaven) and economic mischief for others (those whose jobs are being displaced). And no sector appears to be safe.
So, do robots displace human workers? This turns out to depend on what country you’re talking about—and in most countries, the answer at this point still appears to be no. The FT has reproduced some interesting charts from the International Federation of Robotics, tracking numbers of robots versus trends in unemployment.
In fact, the only country where the number of robots and the unemployment rate have both been rising is the US. Can we generalize here? Probably not. China, Korea and Brazil have shorter histories in manufacturing—and the use of robots in the process—than do the US, Japan and Germany. China itself has seen a huge surge in robots the past several years, so much so that some observers now think that the rapid pace of robot adoption in China will have a negative impact on manufacturing employment there. Again, these are empirical issues.
But an impact there will be, no doubt about that. The FT has a nice quote from economist Robert Skidelsky concerning what may be coming along:
Recently, automation in manufacturing has expanded even to areas where labor has been relatively cheap. In 2011, Chinese companies spent ¥8 billion ($1.3 billion) on industrial robots. Foxconn, which build iPads for Apple, hopes to have their first fully automated plant in operation sometime in the next 5-10 years. Now the substitution of capital for labor is moving beyond manufacturing. The most mundane example is one you will see in every supermarket: checkout staff replaced by a single employee monitoring a bank of self-service machines. (Though perhaps this is not automation proper – the supermarket has just shifted some of the work of shopping onto the customer.)
For those who dread the threat that automation poses to low-skilled labor, a ready answer is to train people for better jobs. But technological progress is now eating up the better jobs, too. A wide range of jobs that we now think of as skilled, secure, and irreducibly human may be the next casualties of technological change.
Skidelsky’s suggestion? Move people to a 30 or 20 hour week. Not a bad idea—I’m certainly looking forward to spending a lot less time what I do for a living than I spend now—I’m past retirement age anyway. But Skidelsky concedes that this is unlikely unless there is some sort of social transformation first—and he doesn’t see that coming along at the moment.
There now appears to be some backlash to the range of happy assumptions about the positive impact of technology—even within the technology sphere itself. Economists have generally relied on the (usually correct) assumption that technological changes, while generating short term losers, generally has had overall positive economic (and therefore societal) effects (although their occasional externalities, like those associated with the internal combustion engine, usually don’t get factored in.) But some economists, including a pair at, of all places, MIT, are starting to question some of those assumptions, especially now, when technology appears to be accelerating and its impacts become more unpredictable. Not that they’re ready to give up the linkage between technology and prosperity that has proved so durable—and it’s not the case that everyone agrees with them. But still, economists Erik Brynjolfsson and Andrew McAfee have clearly touched a nerve when they suggest that not all is well:
Perhaps the most damning piece of evidence, according to Brynjolfsson, is a chart that only an economist could love. In economics, productivity—the amount of economic value created for a given unit of input, such as an hour of labor—is a crucial indicator of growth and wealth creation. It is a measure of progress. On the chart Brynjolfsson likes to show, separate lines represent productivity and total employment in the United States. For years after World War II, the two lines closely tracked each other, with increases in jobs corresponding to increases in productivity. The pattern is clear: as businesses generated more value from their workers, the country as a whole became richer, which fueled more economic activity and created even more jobs. Then, beginning in 2000, the lines diverge; productivity continues to rise robustly, but employment suddenly wilts. By 2011, a significant gap appears between the two lines, showing economic growth with no parallel increase in job creation. Brynjolfsson and McAfee call it the “great decoupling.” And Brynjolfsson says he is confident that technology is behind both the healthy growth in productivity and the weak growth in jobs.
It’s a startling assertion because it threatens the faith that many economists place in technological progress. Brynjolfsson and McAfee still believe that technology boosts productivity and makes societies wealthier, but they think that it can also have a dark side: technological progress is eliminating the need for many types of jobs and leaving the typical worker worse off than before. ¬Brynjolfsson can point to a second chart indicating that median income is failing to rise even as the gross domestic product soars. “It’s the great paradox of our era,” he says. “Productivity is at record levels, innovation has never been faster, and yet at the same time, we have a falling median income and we have fewer jobs. People are falling behind because technology is advancing so fast and our skills and organizations aren’t keeping up.”
And people will likely continue to fall behind, simply because of that troubling problem we mentioned earlier—the long term unemployable. It’s pretty well demonstrated at this point that if you’ve been looking for work for more than six months, you’ve got a problem, at least in the US. And some of this clearly relates to what Keynes, back in the day, referred to as “technological unemployment.” But some of it also reflects the fact that there are more people than there are jobs for them to do. And it’s difficult to see how problems in this area are to be avoided, given current trends. Those kids who used to pump gas are a victim of this, as are those cashiers who aren’t there anymore because of supermarket scanners. As are any number of process control engineers and machine tool operators, which in the past were skilled occupations, now driven towards, if not extinction, then at least a significantly smaller job pool, by advances in technology.
Yes, it’s true that technology can create more jobs—and in the past this has often been the case. Whether we’re at some sort of inflection point on this in the US (and eventually elsewhere as well) remains to be seen. But the technology, as we’ve seen, continues to get more and more impressive in this regard. Just ask the people who used to do the welding on the Ford or GM assembly line. Of course, the role of welding is being obsolesced to some extent as well—as cars continue to contain more synthetic and composite materials (checked out what your bumper is made out of lately?), many of the current generation of welding robots may be out of a job as well. Small comfort, though.
And it is a general trend. It’s been known for a while now, as a recent column by Bruce Bartlett reminds us, that Labor’s share of National Incomes continues to decline. And this is a global phenomenon, not just one associated with the US. And countering this trend in a world dominated by too-big-to-fail banks and companies where CEO pay has increased by 875% over the same 30-year period is going to prove difficult. To complicate the issue further, “Labor” is a bit of an artificial construct in this regard, helpful for statistics but not very good at capturing the real world. Because we’ve seen a bifurcation between high-skilled labor, as Bartlett points out—the people who have been in a position to take advantage of the productivity increases o the past three decades—and low-skilled labor, those people whose textile plant jobs were replaced by workers in China. And the problem seems to be that the path from the latter to the former contains more obstacles—or fewer established routes.
So where does this leave us? Well, in an uncomfortable place, sadly. Because, among other things, we remain surrounded by economists who still don’t get it. The biggest offenders remain those who think that if we bring the debt down, the economy will take care of itself. This is patently absurd, and most policy-makers know this. But we also keep hearing about the need for job growth as if this is a process that can be managed. Well, sure. But where? Even if we really wanted the US economy to resume its dependence on homebuilding as a critical driver of economic growth (not forgetting financial services, of course), which actually doesn’t seem like a good ideal at all when you think about it, you still need the job growth elsewhere so that people can afford those homes. I’m sure McKinsey is correct—there will be 1.6 billion new consumers coming along in coming decades, all of whom will want and need to buy stuff. But that stuff is probably going to be made somewhere else than the US—assuming that the resource base exists to support this.
And those 1.8 billion new consumers? What work will they be doing that will be generating the incomes that will let them make those purchases? And what are their expectations? China is now the world’s largest car market. But can everyone in China own and drive a car? Is this even a remotely good idea? As Lester Brown pointed out nearly two decades ago, there just aren’t the resources for the Chinese to eat like Americans and Europeans. And there are all sorts of other constraints that crop up—perhaps most importantly, water. Still, even if on e assumes that these constraints don’t kick in, what is there for potential workers in the US to do? And not just the US—but places like Russia, France, and Spain, where a large industrial infrastructure just can’t be economically supported according to the current rules of how the world works (although France keeps doing better than it should). Where will the jobs come from? What sort of jobs will they be?
Well, this is where we usually toss out some lofty thoughts about localism and Wendell Berry and the dignity of working with your hands, that sort of thing. And it’s true—what the US needs is a set of local and regional economies that are able to stand more independently than they could at present, if they even existed. This sort of problem arises in Europe only in the larger countries—the UK, Germany, France and Spain. Certainly the latter two have a north/south divide, much as the UK does. But the UK is small geographically, compared with the US. The US is a big country. Which is why any solution involving localism needs to be, well, a whole lot more local. The Transition Towns movement that we have written about before is the sort of thing that needs to be developed—and it can only really be developed locally. It’s designed to be local. And John Robb’s Resilient Communities initiative is very much on the right track. But a move towards localism and regionalism will inherently create regional imbalances—some areas are simply more bountiful than others. New England’s growing season isn’t the same as North Carolina’s. And the southwest? What does it do when the water is gone?
But this is going to require some substantial local, regional and even national recalibration of expectations for jobs and, yes, incomes. In the absence of this, we’re going to continue to see lots of local and regional developments where developments aren’t necessarily benign. Marijuana may or may not be the largest cash crop in the US, but it certainly seems to be a pretty valuable one. And let’s not even talk about meth labs, which have a habit of popping up in places where there’s not much else to do—especially economically.
It’s interesting to think about this issue in the context of how countries have dealt with surplus populations in the past—in the case of Europe and much of Latin America, they’ve sent them to the US. But it’s not clear that this is still a viable option. Yes, there are all those jobs that Americans don’t want to do that immigrants are, well, not exactly happy to do, but they do them anyway. And for the most part these are jobs that someone does need to do. It’s a big world out there. But, still, it seems as if there really are only going to be a finite number of jobs to go around in the manufacturing sector—although lots of jobs at fast food restaurants. We need a complete re-think here. There’s always going to be a need for nurses, and for forest rangers. But the job on the assembly line is going to be evolving—it already has been. And it’s just not going to provide the economic paths that it used to. In fact, it’s becoming increasingly difficult to think that these paths even still exist.
If I were to make a start anywhere, let’s start by eliminating some technology. How about self service gas pumps? New Jersey and Oregon still require gas to be pumped by attendants. Sounds good to me. It would give some kids full or part time jobs that don’t exist now. And my hands wouldn’t smell of gasoline every time I stop in the gas station.