Felix Salmon

It’s long — at some 4,500 words — but I can highly recommend the debate between George Soros and Hans-Werner Sinn about what Soros calls The German Question.

The debate is profound, and the two stake out radically different positions, even though they end up at pretty much the same place. Soros says that Germany should make a simple choice: either sign on to Eurobonds, where the euro zone as a whole would issue low-yielding debt to the benefit of all, or else leave the euro zone entirely. Either way, he says, Europe would win — either from reducing the fiscal burden of the various national debts, or else from seeing the euro devalue against the new Deutschmark.

Sinn agrees with Soros that Germany would be making a huge mistake were it to leave the euro zone; he disagrees vehemently, however, on the subject of Eurobonds. But both men are clear that given political realities in Germany, neither of Soros’s two choices is going to happen. Germany is going to stay in the euro zone, and Eurobonds aren’t going to happen.

That, says Soros, is a tragedy:

Europe would be infinitely better off if Germany made a definitive choice between Eurobonds and a eurozone exit, regardless of the outcome; indeed, Germany would be better off as well. The situation is deteriorating, and, in the longer term, it is bound to become unsustainable…

There is no escaping the conclusion that current policies are ill-conceived. They do not even serve Germany’s narrow national self-interest, because the results are politically and humanly intolerable; eventually they will not be tolerated. There is a real danger that the euro will destroy the EU and leave Europe seething with resentments and unsettled claims. The danger may not be imminent, but the later it happens the worse the consequences. That is not in Germany’s interest.

And even though Sinn thinks that Soros is wrong, his prognosis seems just as grim, filled with painful austerity and sovereign default:

The only remaining option, as unpleasant as it may be for some countries, is to tighten budget constraints in the eurozone. After years of easy money, a way back to reality must be found. If a country is bankrupt, it must let its creditors know that it cannot repay its debts.

My sympathies in this debate are with Soros, although Sinn does make a good point about the unintended consequences of Alexander Hamilton mutualizing state debts in 1791. (There really aren’t a lot of precedents for the kind of Eurobonds Soros envisages.) The overarching message from both of them, however, is that, as Soros puts it, “the current state of affairs is intolerable”. The only question is whether there’s a better alternative; Soros says there is, while Sinn says there isn’t.

The conclusion from them both, then, would seem to be that Europe as a whole is doomed to misery for as far as the eye can see, and that things are going to get worse before they get worse. I really hope they’re wrong. But so long as Europe’s future generations remain jobless, it’s hard to see a silver lining to this cloud.

Today’s jobs report was a solid one, and shows that the recovery, while not exactly strong, is at least not slowing down: Neil Irwin calls it “amazingly consistent”. Whether you look at the past 1 month, 12 months, 24 months, or 36 months, you’ll see the same thing: average payrolls growth of roughly 170,000 jobs per month. That’s not enough to bring unemployment down very quickly, given the natural growth in the workforce. But unemployment is coming down slowly. And at the rate we’re going, at some point in the second half of 2014 we should see total payrolls reach their pre-crisis levels, and the headline unemployment rate hit the key 6.5% level.

There’s a real human cost to the fact that unemployment is coming down so slowly, but there are lots of reasons why it’s very hard to bring it down more quickly. First and foremost, of course, is the fact that US GDP growth is mediocre, coming in at less than 2% per year over the past few years. That’s not the kind of V-shaped recovery which creates jobs. Calculated Risk’s justly-famous jobs chart shows just how bad the recession was for employment, and just how painfully slowly we’re scratching our way back: we’re more than five years into this jobs recession, and we’re still at the worst levels seen in the wake of the dot-com bust.

One of the reasons is the undisputed conclusion of Reinhart and Rogoff: that recoveries from financial crises are much slower than recoveries from other crises. But there’s something bigger going on, too, which Joe Stiglitz writes about today in a very wonky blog post for the IMF.

This is more than just a balance sheet crisis. There is a deeper cause: The United States and Europe are going through a structural transformation. There is a structural transformation associated with the move from manufacturing to a service sector economy. Additionally, changing comparative advantages requires massive adjustments in the structure of the North Atlantic countries.

To put it another way: what looks like a broad economic recovery is actually a combination of many trends, including the end of what turned out to be a very short and weak recovery in manufacturing employment. Here’s Irwin:

The fact that the overall job growth numbers have been extraordinarily stable does not mean there isn’t some real churn going on in the U.S. workforce. In the earliest phase of the recovery, manufacturing jobs was a major driver of job creation, but that turned out to be not a longer-term trend but a partial reversal of the steep declines of the recession. Now, job creation is entirely confined to the services sector: Manufacturing had no net change in employment, construction lost 6,000 jobs, and even mining and logging was a net negative.

Government employment, meanwhile, continued its long swoon… That leaves one sector to drive the train of job creation: private sector services. This particular month, there were strong gains in leisure and hospitality, retail jobs, and professional and business services, and health care has been a mainstay of the expansion.

Stiglitz makes the case that in a recovery with so many moving parts, the single blunt instrument of setting short-term interest rates at the Fed will never be enough, and that “there needs to be close coordination between monetary and fiscal policy.”

What’s more, as Mohamed El-Erian says, policymakers should ideally be able to use job growth not just as a goal, but also as a tool for achieving other ends.

Robust employment growth would – and, let us hope, will – play a critical role in helping the US pivot to a better place… It would do this by maintaining consumption and allowing for a more sustainable savings rate; by countering an excessive upfront fall in public spending that increases the risk of a recession; by enabling the Fed to slowly and gradually normalise monetary policy before it breaks too many things; and by reducing the risk of financial bubbles.

The US economy is a highly complex machine, with many moving parts which ought to be working with each other rather than against each other. Stiglitz makes a strong case that the financial sector broadly is right now part of the problem rather than part of the solution: it’s not directing funding to help the economy grow and create jobs, even as it continues to represent a serious systemic risk. It should go without saying at this point that fiscal policy broadly is part of the problem as well: you don’t create jobs by firing people, and the government should be borrowing if and when the private sector won’t. And as for monetary policy — well, it’s probably too early to tell. It’s done a great job of making people with money richer, but it has had a much less obvious effect on creating jobs for those who want them and don’t have them.

And yet there’s real room for optimism in today’s jobs report. Look at the revised numbers for February: an incredibly heartening 332,000 jobs created, in one short month. Look at the number of people unemployed for 27 weeks or more: that unhappy cohort shrank by 5.6% in April alone, to 4.3 million people. It’s still far too high, but this time last year it was over 5 million, so we’re making a significant dent in what has been the toughest nut to crack.

We can — and should, and could, and must — do better than this. But doing so will require a thaw in the Washington gridlock. When Jack Lew became Treasury secretary, it was understood that the most crucial thing he could deliver would be greater cooperation between the White House, Treasury, and Capitol Hill. That hasn’t happened yet. I hope and trust that he’s been working very hard behind the scenes to make it happen — partly because he doesn’t seem to have achieved anything else, but mainly because it’s by far the most important thing that he could be doing right now. Behind the jobs numbers there are some powerful forces driving real recovery in large parts of the US economy. It’s Lew’s job to work with Congress to identify those forces, and to give them all the support the government can muster.

Paul Krugman and Megan McArdle both point to this chart today:

What you’re looking at here is the plight of the long term unemployed in the wake of the Great Recession. If you look at the economy before the recession (the blue line), it works pretty much as you think it would: as the number of job openings goes up, the long term unemployment rate goes down. But then the crisis happened, and now we’re in a Bizarro world where the long term unemployment rate goes up even as the number of job openings increases.

It’s worth looking at this chart in the context of one which might be more familiar:

What you’re looking at here is initial claims for unemployment (the blue line) and the unemployment rate (the red line); both are rebased to 100 at the end of 2007. You can see that initial claims have historically been a very good leading indicator when it comes to the unemployment rate. And that’s perfectly intuitive: if the number of people newly claiming unemployment each week is going down, you’d expect the overall unemployment rate to follow.

But there’s something worrying about this chart: although the unemployment rate is indeed coming down, it’s not coming down as fast as you’d expect it to, given the sharp drop in initial unemployment claims. In other words, people aren’t becoming newly unemployed, but the unemployment rate is still staying stubbornly high. Which is another way of saying that this time around, the long-term unemployed are finding it particularly difficult to get back to work.

I decided to put together the exact same chart, only instead of using the overall unemployment rate, I’d look at just the long-term unemployment rate — the proportion of people who have been unemployed for more than 27 weeks. This is what I found:

The blue line, in this chart, is exactly the same as the blue line in the chart above it. But the red line is long term unemployment — which is at massively unprecedented levels.

This chart tells me two things. Firstly, it is indeed the long-term unemployed who are the reason why the unemployment rate overall isn’t coming down as fast as it should be. And more importantly, there’s a quiet humanitarian disaster happening right under our noses. Here’s McArdle:

Short of death or a debilitating terminal disease, long-term unemployment is about the worst thing that can happen to you in the modern world.  It’s economically awful, socially terrible, and a horrifying blow to your self-esteem and happiness.  It cuts you off from the mass of your peers and puts stress on your family, making it likely that further awful things, like divorce or suicide, will be in your near future.

McArdle and Krugman differ on the policies that should be enacted to address this emergency — but they agree that policies should be enacted to address this emergency, with urgency. That’s where they both part ways with Congress, which is much more interested in deficit reduction than it is in trying to make a dent in the long term unemployment rate.

The lesson of the past few years is that this is not a normal recovery: corporate profits are doing great, while total employment remains anemic. We can’t trust the invisible hand to generate the millions of jobs that are needed, especially with regards to the long-term unemployed. With gridlock in Washington, the result is a huge amount of unnecessary human misery.

Today’s jobs report is an unambiguously positive one: America had 236,000 more jobs in February than it had in January, and the unemployment rate is down to 7.7%, the lowest it’s been since 2008, before Barack Obama was even sworn in. (Although, it’s still nowhere near the 6.5% at which the Fed will start thinking about tightening monetary policy.) Things are getting better, US fiscal policy notwithstanding, and it’s great to see construction in particular, especially non-residential construction, finally making a substantial positive contribution to the numbers.

All is not entirely sweetness and light, though, as Brad DeLong and many others have noted. The number of multiple jobholders rose by 340,000 this month, to 7.26 million — a rise larger than the headline rise in payrolls. Which means that one way of looking at this report is to say that all of the new jobs created were second or third jobs, going to people who were already employed elsewhere. Meanwhile, the number of people unemployed for six months or longer went up by 89,000 people this month, to 4.8 million, and the average duration of unemployment also rose, to 36.9 weeks from 35.3 weeks.

In terms of the economy, it’s not good enough to simply increase employment and decrease unemployment, if the proportion of people with jobs isn’t actually going up. Which is why this chart, from Calculated Risk, is the most important one to look at right now:

Both the employment-to-population ratio ad the labor force participation rate are much lower than they ought to be: if this is a recovery, the former in particular ought to be going up, rather than going nowhere. Yes, it’s important to ensure that the unemployed get jobs. But in many ways it’s even more important to try to create jobs for people who simply aren’t working, rather than just for the people who are actively looking for work.

To turn these ratios into hard numbers: there are 89.3 million Americans who are not in the labor force, of whom just 6.8 million currently want a job. The economy ought to be able to find good, rewarding jobs not only for the 6.8 million, but for a large chunk of the other 82.5 million as well. Just imagine what that would do for tax revenues: all our fiscal problems would be solved at a stroke!

This month’s payrolls report is one only a statistician could love. The official press release contains three whole pages on the subject of revisions and rebenchmarking, on top of the box on the front page: the very clear message is that we’re starting a new year, and that it’s a very bad idea to compare the January numbers to the December numbers.

That said, no one would have been particularly surprised by these numbers even if they were directly comparable: they’re basically exactly in line with what we’ve been seeing for months. Payrolls growth is constant, unemployment is flat: nothing’s getting better, nothing’s getting worse. And if you exclude some of the more volatile categories, that trend is even stronger. The good news in the report is about levels rather than changes: the total number of Americans on payrolls is now 134,825,000, while last month we were told that it was just 134,021,000. That means the number of Americans with a job is 800,000 higher than we thought last month, even if the unemployment rate, at 7.9%, is no closer than it was in September to the Fed’s 6.5% target.

Frankly, the payrolls report should be of interest mostly to statisticians and econometricians. The error bar for the number of jobs added is large, but the amount of data in there is enormous: the jobs report is an amazingly useful resource for people who want to study the state of the US economy over time. The problem is that it’s also the earliest indication that the markets get of how the economy is doing each month, and as a result it moves markets more than any other scheduled data release in the world, with the possible exception of FOMC announcements. In our short-attention-span world, where anything that moves markets must be news, it’s far too easy to get distracted by monthly noise — especially when people like me feel the need to weigh in every month.

The fact is that the jobs report is a sprawling, messy thing at the best of times, and it’s especially so this month. We only ever find out with hindsight whether the immediate market reaction was rational or silly, and the same goes in spades for the flurry of activity that greets the report on Twitter every month. Even the unemployment rate, which is now being explicitly targeted by the Fed, isn’t crucially important: Ben Bernanke has made it very clear that he’s going to take a big-picture view of the employment situation to determine when it’s time to raise interest rates, rather than just target the U3 rate narrowly, with no attention paid to things like the workforce participation rate.

To be honest, this month’s payrolls report probably isn’t even the most important data release of the day, let alone the month: the data from the manufacturing sector of the US economy is much less ambiguous, with the ISM report coming in strong and GM sales looking impressive. It’s an open question, of course, as to whether and how industrial strength is going to make its way through into full employment. But don’t look to the headlines atop this month’s payrolls report for answers. If they’re in there at all, they’re deeply buried, and not easy to find.

This is the US unemployment rate, from Calculated Risk. Today’s jobs report was a very positive one: not only did job creation exceed all expectations, but unemployment fell too, to 7.7%. For the first time, the unemployment rate is lower than it was when Barack Obama took office, in January 2009.

The employment recovery is now 33 months old, and as strong as it’s ever been. We’re still a long way from achieving pre-recession levels of employment, but the fact is that it’s hard to maintain a sense of crisis and emergency for this much time: if you live with anything for more than a couple of years it becomes normal. (Which is one reason why Europe, which has a structurally much higher unemployment rate than the US, doesn’t consider itself to be in a permanent jobs crisis.)

The levels in the employment report are still scary. 7.7% is high in absolute terms, and both the employment-to-population ratio and the labor force participation rate are much lower than they should be. America should have millions more people at work than it does, and there’s a very strong case, looking at levels alone, for further economic stimulus to help us further in the right direction.

But there’s something oxymoronic about the concept of a permanent state of emergency. And in terms of how strong the recovery feels, first derivatives are just as important as levels: if unemployment has fallen from 8.7% to 7.7% in the past year, that feels better than an economy where unemployment has risen from, say, 6.1% to 7.1%. When the temporary payroll tax cut was passed, unemployment was higher than it is now, and it was rising; clearly we’re in a much better spot now than we were then.

The best-case outcome from the fiscal negotiations now taking place between Barack Obama and John Boehner is that they move us out of the “permanent temporary” tax code and into a world where everybody knows what tax rates are and what they will be. Putting expiry dates on tax cuts is a gimmick, and while there’s a case for doing that kind of thing in the middle of a major crisis, we’re really not in the middle of a major crisis any more. It took far too long for the unemployment rate to start falling, and it has been falling far too slowly. But “unemployment should be falling faster” is not a crisis.

With any luck, then, the resolution to the fiscal-cliff debate will be a set of tax policies that both sides agree on, along with a clear date when they will be fully in force. I’m thinking January 1, 2014. The key number to look at will be total federal taxes as a percentage of GDP: it needs to be high enough to be able to run a mature modern democracy. Then, once you have a clear and permanent tax code as your primed canvas, you can start having a sensible conversation about government expenditures: where they need to come down, and which areas of the economy need some stimulus. Even if spending-related stimulus is no more effective than tax-cut-related stimulus, it’s still a better option, because it allows you to leave the tax code alone.

If Obama’s first term was about doing whatever was necessary to get us out of the biggest crisis in living memory, his second term should be dedicated to building strong and permanent foundations for the economy going forward. America’s fiscal architecture is a key part of that — indeed, it’s the key part. So if the payroll cut disappears, along with all other temporary bells and whistles, that’s fine. What’s good for the economy now will also be good for the economy next year, and the year after, and the year after that. Let’s structure any a deal so that it can work forever. And then, if there are temporary political and economic issues which need addressing, let’s tackle them through means other than the tax code.

Charles Fishman has an upbeat cover story in the new Atlantic, talking about how manufacturing is making its way back from China to America. As the world demands an ever-more nimble manufacturing sector, able to produce smaller quantities of goods more quickly, it makes sense to make those goods here rather than be forced to spend a month shipping them over from China, especially with shipping costs rising. On top of that, Chinese manufacturing costs are rising too: inputs from labor to natural gas are getting much more expensive. (Natural gas costs four times as much in China as in the U.S., while James Fallows reports that a typical Foxconn salary is now $400 a month, three times what it was six years ago.)

Fishman’s enthusiasm for bringing the designers closer to the means of production — it really does make for much more efficient assembly lines — means that he papers over the reality of what America’s new manufacturing-sector workers are being paid:

Appliance Park’s union was so fractious in the ’70s and ’80s that the place was known as “Strike City.” That same union agreed to a two-tier wage scale in 2005—and today, 70 percent of the jobs there are on the lower tier, which starts at just over $13.50 an hour, almost $8 less than what the starting wage used to be.

There’s a huge difference between $13.50 per hour and $21 per hour: the latter is something you can actually live on, something you can consider to be a career. The former is not. And that’s a problem, as Adam Davidson explains: the reason that people aren’t going to college to learn the skills needed on a modern manufacturing assembly line is simply that those skills aren’t valued highly enough. Even McDonald’s, where there were noisy strikes today, looks attractive in comparison:

Isbister doesn’t abide by strict work rules and $30-an-hour salaries. At GenMet, the starting pay is $10 an hour. Those with an associate degree can make $15, which can rise to $18 an hour after several years of good performance. From what I understand, a new shift manager at a nearby McDonald’s can earn around $14 an hour.

What we’re seeing here is the same thing that Seth Ackerman saw at Hostess: wages that are so low, the workers prefer to give up the work entirely, and take a better-paying job elsewhere.

In a piece for Salon, Jake Blumgart quoted a bakery worker who had been at the company for 14 years. “In 2005, before concessions I made $48,000, last year I made $34,000…. I would make $25,000 in five years if I took their offer. It will be hard to replace the job I had, but it will be easy to replace the job they were trying to give me.”

What we have here is a situation where a company offered a wage in the marketplace and couldn’t get any workers to accept it. Consequently, it went out of business. The word “competitive” gets thrown around a lot, often with the murkiest of meanings, but in this case there can be no doubt at all that a company, Hostess, was unable to pay a competitive wage. Ninety-two percent of its workers voted to walk out on their jobs rather than accept its wage, and they stayed out even after they were told it was the company’s final offer.

All of which means that there are two enormous problems with the story that manufacturing is returning to the US. That might be true, but (a) it’s not creating many jobs, and (b) the jobs it is creating are not the good jobs which people want to have for many years. Instead, they pay $15ish per hour, which is what teenage babysitters make in New York.

Once upon a time, in the halcyon 1950s and 1960s, a man could have a blue-collar factory job and make enough money to support a whole family. Those days are over now, but they echo still in the dreams of manufacturing returning to the U.S. The idea is that were that to happen, good jobs would magically be created. Where the reality is that manufacturing jobs are not good jobs any more: you’re better off working in retail, whether you’re in the US or in China. And you don’t need to spend unpaid years in college learning technical skills to get a retail job.

So while I’m as excited about the Internet of Things as the next guy, and I love any economy where ideas can become products with unprecedented ease, I don’t think that this is a particularly good solution to the unemployment problem. It’s better than nothing, of course. But I do get worried when The Atlantic splashes the word “COMEBACK” all over its cover: that makes this phenomenon seem much happier than in truth it is.

There’s lots of good news in today’s employment report: payrolls rose substantially in October, and the already-great numbers for August and September were revised upwards to boot. Even the uptick in the unemployment rate, from 7.8% to 7.9%, was actually positive in many ways. Americans are back looking for work, which bodes well for the next few months.

And then there’s the technocratic good news, too: the BLS isn’t just releasing numbers any more, it’s also releasing charts! Along with the standard payrolls press release this morning, there was also a PDF file which included the chart above.

This chart I think tells the big-picture story very well. Firstly, jobs aren’t political, they’re economic. They rise when the economy is doing well, and fall during recessions: the name or party of the president can only affect things at the margin.

Secondly, we’re still in very bad shape, employment-wise: there’s a long way to go before we get back to where we were before the crisis. And that’s in absolute terms: remember that the US population has been growing all along.

And thirdly, in case you needed any reminder, this recovery is long and painful. Look at the rate of employment growth from 2002 to 2006, and extrapolate it upwards to get an idea of what the capacity of the US economy is: where we could be, if we hadn’t been hit by the crisis. Then, look at the gradient of the current recovery: it’s not noticeably steeper than the trend-growth gradient. Which means, to a first approximation, that we’re just as far below capacity as we were when the recession ended.

This isn’t entirely bad news. For one thing, the recession has ended, thanks in no small part to the 2009 stimulus and to the unprecedented monetary operations being carried out by the Fed. And compared to, say, any country in Europe, the strength of the US recovery is decidedly impressive. But if you aspire to full employment — and that is one of the Fed’s two mandates, after all — then in one sense we’re just as far away from that goal as we were two years ago.

Markets and economists will and should react positively to today’s report, which is significantly better than most of us expected. But it doesn’t change the fact that the biggest problem facing the US economy is unemployment and underemployment. We need to get the unemployed and underemployed working again, and the longer it takes to do that, the harder the job becomes, given the well-documented ways in which long-term unemployment erodes skills and morale.

How much can politicians really do, on that front? I don’t know. But a front-loaded fiscal employment push would be great right about now, while massive spending cuts are precisely the opposite of what we need. And in a country where millions daily face the misery of being unable to find work, it would really be unconscionable to vote for anybody who thinks it would be a good idea to cut food stamps by $134 billion.

Every little vertical notch on that BLS chart represents 2 million employed Americans. We all want to see the lines go up and to the right, so that millions more Americans get jobs. But we also need to keep in mind just how depressed those lines are. The number of Americans without work is absolutely enormous right now. And so while we’re trying to create new jobs, it’s even more important that we ensure the well-being of those who don’t have employment yet.

Last month, I wrote a bit about what you might call the bidirectional causality between the unemployment rate and incumbent political fortunes. On the one hand, unemployment affects happiness, and willingness to vote for the current president. But on the other hand, the rate itself is a political weapon, to be used in, for instance, Republican claims of how many successive months America has had unemployment above 8%.

This morning’s job report is fascinating, in that it illuminates both effects. One not-bad way of judging Obama’s odds of winning the election is to look at his odds on InTrade, which were stubbornly close to 60% for pretty much all of this year. Then, suddenly, in mid September, there was a huge surge. Narrowly speaking, the surge was a result of Obama doing well in important polls. But why were the polls suddenly turning so much in Obama’s favor? Journalists, with their bias towards believing that everything is a result of news, tended to attribute Obama’s newfound popularity to Romney gaffes like his 47% speech. But with hindsight, maybe news had very little to do with it. Maybe the real reason was that according to the household survey, the number of employed Americans grew by 873,000 in September alone.

Which isn’t to say that news is irrelevant when it comes to perceived election odds. For instance, within the first half-hour of the debate on Wednesday, Obama’s InTrade odds fell from 71% to 67%, just on the back of his weak performance. And then, this morning, they spiked right back up again, from 65% to 71%.

This move is about optics as much as it is about reality: with the unemployment rate now below 8%, a key Republican talking point has been neutralized. And for the wonkier types, Obama can now say that he’s created more private-sector jobs in the past four years than George W Bush created in eight. Indeed, if it weren’t for public-sector job losses — exactly the kind of spending cuts that Republicans claim to love — the unemployment rate right now would have a 6 handle.

Those optics explain the frantic and ignoble conspiracy theories from the Republican side: they’re trying to alter perceptions of the number itself, even if they can’t alter the effect that rising employment has on the electorate’s propensity to vote for Obama. Because it seems as though rising employment is giving a significant boost to Obama’s re-election chances — and that from here on in, it will only be helped by the 7.8% headline unemployment rate. It’s a little bit depressing that 7.8% counts as low, for these purposes, but clearly it does — especially considering that it has come down 1.2 percentage points in the past year. I don’t know how much credit Obama can really take for that, but America, right now, seems to be willing to give him the benefit of the doubt.

Tim Kane, at the Hudson Institute, has a new paper out with a simple title: “The Collapse of Startups in Job Creation”. His paper is basically a slightly politicized version of the charts put out by the Bureau of Labor Statistics last month, under the headline “Entrepreneurship and the U.S. Economy”. The first two charts are particularly striking. The first one looks at the number of startups in America — companies less than one year old.

This shows a reasonably steady rise in entrepreneurship from 1994 to 2006, then a collapse as the housing bubble bursts, and — most worryingly of all — no recovery at all after the recession ends. Instead, we have significantly fewer startups right now than we did even at the depths of the recession.

If you look at the number of jobs at these startups, rather than the number of startups, the picture is equally bad, although the decline is older. This series peaked back in 2000, and has been declining ever since:

This doesn’t make a lot of intuitive sense. As Kane writes,

Economic theory suggests that the modern economy offers a better environment for even more entrepreneurship. First, there is a wider technology frontier to explore. Second, a wealthier society enables more individuals to explore rather than merely work to survive. Third, the shift to services requires less startup capital than manufacturing or agriculture. In other words, the downward trend in the rate of entrepreneurship should, in theory, have rebounded by now.

Kane thinks that it’s something to do with taxes and regulations; I don’t buy it. But he also has a globalization argument:

An American entrepreneur has zero tax or regulatory burden when hiring a consultant/contractor who resides abroad. But that same employer is subject to paperwork, taxation, and possible IRS harassment if employing U.S.-based contractors.

Are jobs at US startups effectively being offshored? I don’t know. But I do know that small business is where the jobs are, in this economy. Here’s the chart:

The green line, at the top, is the number of jobs at small businesses, with less than 50 employees. The red line, underneath it, is the number of jobs at medium-sized businesses, with somewhere between 50 and 500 employees. And the steadily-declining blue line, at the bottom, is the number of jobs at large businesses with more than 500 employees. Clearly, if we want to boost job creation, the best place to look is not the blue line but the green line. And equally clearly there has been an increase in the number of jobs at small firms overall, since the recession ended.

So if small firms in general are hiring again, what’s the problem with startups? Kane has run the numbers back to 1989, to come up with this chart:

There’s really nothing predictable about the dismal showing in the last three years of this chart — and especially not in the last two years, when we’ve had a recovery accompanied by record-low interest rates.

Admittedly, all of these numbers are low: at their peak, startups employed only a little more than 1% of the population, and now they employ a little less than 1% of the population. Concentrating on startups is not going to move the broader employment needle very much. But the dynamic here is surprising and troubling, all the same. Intuitively, if people can’t find work for an existing company, they should be more likely, not less likely, to go out and found a new company themselves, instead. But that doesn’t seem to be happening.

The only thing I can think of here is that for all that we think of startups as being largely high-tech things, in reality a huge number of them are in the construction industry, in one way or another. In a word, subcontractors. And no one’s starting new granite-countertop installation companies right now. But still, startups are a decent proxy for the dynamism of an economy. And these charts don’t bode at all well, on that front.

First, go and check out Mike Konczal’s wondrous gift to the internet today, a fabulous GIF-filled guide to QE3 and monetary policy, where I found the GIF above. It neatly encapsulates the greatest problem facing America today: people can’t find work.

This is not an existential crisis on the order of the financial crisis of 2008-9: if we fail to solve it, the entire economy won’t grind to a halt. But it’s a crisis all the same, and it’s being tackled with much less urgency — and much less money — than was brought to bear on the financial crisis.

And so Deborah Solomon’s idea is, at least in principle, a good one. She notes that no one expected Treasury to recoup much if any of the money it spent on the various bailouts, and that TARP, in particular, was considered a permanent government expenditure rather than some kind of temporary loan to the financial system. Now that hundreds of billions of dollars of TARP funds have found their way back to Treasury, couldn’t they be recycled to help, not Wall Street this time, but the unemployed?

The four-year anniversary of the financial crisis should prompt policymakers to consider using some of the money — funds that were never expected to be returned to the government’s coffers — to assist those in need. Some 12.5 million Americans remain unemployed, and 40 percent have been out of work for six months or longer. Almost 15 percent of Americans rely on the government for food stamps. The 8.1 percent unemployment rate, while far better than its peak of 10 percent, crept down last month largely because so many people have given up looking for work.

That $353 billion could go a long way toward helping struggling Americans who would benefit from some type of relief — whether through tax cuts or the extension of benefits. The U.S. taxpayer stepped in to help Wall Street in its time of need. It’s time for Wall Street, albeit indirectly, to return the favor.

The distinction to bear in mind here, however, is not that between Wall Street and Main Street, but rather that between stocks and flows. TARP wasn’t a new income stream for Wall Street; it was a one-off injection of capital, which got Wall Street back onto its feet, to the point at which banks were pretty much the first companies to start becoming hugely profitable again after the recession hit.

So the condign use of TARP funds in an unemployment context would not be through things like tax cuts or benefit extensions. Instead, it would be a massive capital expenditure: we would build a job cannon, and a whole forest of jobies, and manufacture a huge pile of job helmets, and fire off the unemployed to the wonderful world known as jobland, where they could live happily ever after without any further government support.

Buying a bunch of long-dated agency securities — quantitative easing — is not that. Besides, we’re talking about fiscal policy, not monetary policy, here. But even fiscal policy can’t easily create permanent jobs with one-off expenditures. A dynamic economy is just that: it’s dynamic, and the number of people being hired and fired every week is enormous. The total number of jobs is going up, far too slowly. But permanent jobs don’t — and shouldn’t — exist. What we really want to create is an economy where for every person who gets fired, two people get hired. That would bring unemployment down sharply.

Which brings me to the major speech that Barack Obama gave on September 8. No, not his acceptance speech at the Democratic National Convention — that was September 6, in any case. But rather the speech he gave on September 8 2011, presenting — and costing out — the American Jobs Act. And it turns out that the cost of Obama’s jobs act, which he doesn’t seem to mention much any more, was $447 billion: very much in Deborah Solomon’s ballpark.

Would the jobs act create job cannons and job helmets and jobies? Of course not: there’s no simple Main Street corollary to TARP which can be implemented with the requisite political will and a few hundred billion dollars. Job creation is more of an art than a science, and there aren’t any real experts in it. But at least Obama has (or had) a real, detailed plan. And, as Treasury officials love to say, as a general rule, plan beats no plan.

There are two ways in which the national employment situation influences the election. The first is, simply, the effect of unemployment and underemployment on America’s animal spirits. People who are unemployed, or who are so discouraged that they’re not even looking for work any more, don’t tend to be very happy with their lot, and as a result are more likely to vote against the current president. It may or may not be fair, but the president does get blamed for current economic conditions, and arguments about first derivatives (“it’s bad, but it’s getting better”) or counterfactuals (“it’s bad, but it’s better than it would have been under the other guys”) tend to be pretty unpersuasive to voters.

On this level, today’s employment report is pretty gruesome. According to the establishment survey, employers added just 96,000 jobs this month — less than the amount needed just to keep up with population growth. According to the household survey, the size of the civilian labor force shrank by 368,000 people last month. And the number of people not in the labor force grew by an absolutely massive 581,000.

Right now, the proportion of Americans with a job is lower than it has been in over 30 years. America’s getting older, and you’d expect the number to be falling — but it shouldn’t be falling nearly as fast as this. We’re well below trend, when it comes to the employment-to-population ratio, and that’s really bad for the economy as a whole: it means we have fewer productive workers, and as a result the country is creating much less wealth than it could be creating if more people had jobs. At the margin, of course, anything that depresses the amount of wealth in the country is bad for the incumbent president.

So anybody trying to use the jobs report to handicap the result of the election should probably see a tick down, this morning, in the chances of Obama’s re-election. My feeling is, however, that the size of the tick is likely to be very small. These things depend much more on levels than on deltas, and in any case the current electorate is more polarized than ever, with political convictions which are hard to shake.

Which brings me to the second way that employment affects electoral outcomes. The employment numbers are reported, on the first Friday of every month, and political parties try to use the numbers to their best advantage. On this front, there’s really only one number that matters, and that’s the headline unemployment rate. Financial types care more about the payrolls number, because it’s more accurate and less fuzzy. The unemployment rate, by contrast, is harder to calculate, and is based on the idea that you’re only unemployed if you’re looking for work. But the fact is that from a rhetorical perspective, the unemployment rate is the thing which counts. And so in terms of the optics of today’s report, it’s good for Obama, just because the unemployment rate fell — to 8.1% this month from 8.3% last month and 9.1% a year ago.

That’s still well above the 7% at which the psephologists will tell you that it’s very hard for an incumbent to get reelected. And it still starts with an 8 — although there’s now a small chance that on the day we actually vote, the unemployment rate might start with a 7. But the Republicans can’t say that the unemployment rate is rising, and the Democrats can say that it is falling. Will that change votes? Again, not very many. But insofar as arguments have an effect on elections, this report — bad though it is — has failed to give the Republicans the kind of rhetorical ammunition they might have hoped for.

Underlying both of these dynamics is the way in which the story of discouraged workers — people falling out of the labor force entirely — has become increasingly important, to the point at which it makes the headline unemployment rate much less useful as an economic indicator. Once upon a time, if you didn’t have a job, you fell into one of two categories: either you didn’t want to work, or else you were looking for work. Nowadays, however, there’s a huge third category of discouraged workers who would love a job but don’t even see the point of looking any more.

The Bureau of Labor Statistics has an interesting-if-obscure data series called “labor force status flows”. Most of the people interviewed in the survey measuring the unemployment rate, it turns out, were also interviewed the previous month. So it’s possible to look at the number of people, on a month-to-month basis, who were unemployed last month and who were no longer in the labor force this month. Historically, that number has been somewhere between 1.5 million and 2 million per month, on a seasonally-adjusted basis. But when the recession hit, it spiked to more than 2.5 million, and even more than 3 million at the peak. And it’s still extremely high.

It’s natural for lots of unemployed people to move out of the labor force each month: the Boomers are retiring, after all. But a glance at this chart is all it takes to see that we’re well outside normal territory, and that we’re still seeing millions of people leave the labor force not because they want to but because they feel that there’s simply no point in looking for work any more. I don’t know when or whether this line will come back down to its historical levels. But so long as it’s as elevated as this, the Federal Reserve has its work cut out. Because it means that even if the unemployment rate comes down substantially, we still won’t have really reached full employment — not unless the size of the labor force increases substantially at the same time.

It’s the first Friday of the month, which means there’s a new jobs report out, and an absolutely enormous number of people are puzzling over what it might mean. And this close to a general election, there’s also lots of politicians trying to make hay out of it. Both of which, for the intellectually honest, are largely futile exercises at the best of times, and especially so this month.

There are two components to the jobs report: the household survey, which polls households to measure how many people are unemployed; and the establishment survey, which polls employers to measure how many people have jobs. The establishment survey is the more accurate of the two; the household survey is the one politicians gravitate towards. Taken as a package, the two together sometimes convey substantive information on how the economy is doing, and the speed with which we are approaching — or moving away from — full employment.

This month, however, not so much. And in general, the amount of useful information in the monthly jobs report is much lower than the amount of attention paid towards it would imply. Every month, policy wonks and data-heads pore through the 25 different tables in the report, sometimes looking at their own favorite number (U-6 is particularly popular these days), and sometimes just looking for something to jump out at them. (Look at what happened to youth unemployment among Hispanics!)

The latter exercise is particularly dangerous. When you have 25 tables, each with hundreds of datapoints, it’s a statistical certainty that one or two of them are going to be weirdly off in some way. If you then pull out that datapoint and read too much into it, you’re guilty of a logical solecism, a bit like two people who discover they share a birthday and then exclaim at how unlikely that is.

But even the simple check-out-this-headline-number exercise, which every major news organization goes through every month, is statistically dubious. The headline payrolls number has error bars which are more than 100,000 people wide in either direction: if the number everybody’s quoting is 163,000, for instance, all that really says is that there’s a 95% chance that it’s somewhere between 59,000 and 267,000. What’s more, there’s a good 5% chance that it’s outside that wide range. Take the last two years of payrolls headlines, and the chances are that somewhere in there, a number is more than 100,000 off.

And the unemployment rate is less accurate still. Most people will say that it ticked up this month, to 8.3%; the official news release is closer to the mark in saying that it was “essentially unchanged”. And if you look at the numbers it’s based on, they’re all over the place.

This month, the two parts of the report tell differing stories: the headline payrolls number was higher than most people expected, even as the headline unemployment rate went up. Look a bit more deeply, and things become ever messier, what with all the revisions and the changes in people looking for work, and the decrease in median unemployment duration, and so on and so forth. As a result, this month of all months it’s actually quite easy for the markets (if not the politicians) to move on to the next thing. After all, they have very short attention spans at the best of times.

But the fact is that even when the stars align, and the two parts of the jobs report tell the same story, and there seems to be a clear message shining through — even then, there’s nearly always much less to the report than meets the eye. All those numbers will be revised in subsequent months, and the message is never as clear as the media and the markets make it out to be, and in any case the chances are that we’re not actually seeing signal in the noise at all, but instead that we’re just being fooled by randomness.

So be happy, this month, that you can’t work out what the jobs report is trying to say. Because that’s probably exactly the right conclusion to draw on the first Friday of every month.

Michael Kinsley tackles outsourcing today, complaining that Barack Obama is a protectionist who doesn’t understand its value, and that Mitt Romney is keener to pander to protectionists than he is to defend free-market principles. He writes:

Romney or Obama? “I don’t want the next generation of manufacturing jobs taking root in countries like China or Germany.” Early in the Republican primary campaign, China was the one subject Romney seemed genuinely agitated about. Imposing tariffs on Chinese goods was on the long list of things Romney said he was going to do on Day One of his presidency. Maybe he still is, but he doesn’t play it up the way he used to.

Meanwhile, if Romney is a free trader at heart, faking a bit of protectionism, Obama seems to be a protectionist at heart, faking a belief in free trade. That quote in the previous paragraph is from Obama, and shows a fundamental misunderstanding of how markets work. Trade is not a zero-sum game. There isn’t a certain number of manufacturing jobs that will either go to China or Germany, or come to us. We want China and Germany to have lots of manufacturing jobs. The more they have, the richer they are, the better off we will be as well. Beggar-thy-neighbor policies don’t work.

Kinsley is probably right on the politics, here, but he’s wrong on the economics. Here’s Obama’s quote, in context:

I was able to sign trade agreements with Korea and Colombia and Panama so our businesses can sell more goods to those markets. That’s why I’ve fought for investments in schools and community colleges, so that our workers remain the best you’ll find anywhere, and investments in our transportation and communication networks, so that your businesses have more opportunities to take root and grow.

I don’t want America to be a nation that’s primarily known for financial speculation and racking up debt buying stuff from other nations. I want us to be known for making and selling products all over the world stamped with three proud words: “Made in America.” And we can make that happen. (Applause.)

I don’t want the next generation of manufacturing jobs taking root in countries like China or Germany. I want them taking root in places like Michigan and Ohio and Virginia and North Carolina. And that’s a race that America can win.

This, it seems to me, is an entirely coherent economic policy if what you’re trying to do is maximize the number of good working-class jobs in America. There’s no doubt that US employment, as a whole, is on a long-term secular trend away from goods and towards services. And at the same time, the two countries with the world’s biggest trade surpluses — China and Germany — are precisely the two countries with the healthiest manufacturing industries. What’s more, neither of them is suffering a jobs crisis.

So the question arises: should the US continue to accept the Ricardian bargain whereby it concedes to China and Germany the comparative advantage in manufacturing, while keeping for itself the comparative advantage in borrowing-to-consume and constructing synthetic CDOs? The answer is no, and not only because there’s something hollow and dangerous about an economy which is too reliant on financial whiz-bangery. There’s something more important at stake here, and that’s employment.

US manufacturing in fact is extremely competitive on a global scale; the problem is that output has lagged productivity improvements, with the result that we’re making more stuff with ever fewer people.

There’s no particular reason why that should be the case: when manufacturers in China and Germany become more efficient, that’s their sign to employ more people, rather than fewer. As each employee becomes increasingly profitable, it makes perfect sense to keep on adding more employees. Or at least it does in some countries. In the US, by contrast, capital is cheap and plentiful, and there’s much more incentive here to replace people with capital goods wherever possible.

But at that point, why even invest the money in the manufacturing industry at all? Everything becomes a question of opportunity cost, and if you can get higher returns in say the financial sector, then the rational thing to do is to start an investment bank, make lots of money from your trading desk, and then take the proceeds and spend them on manufactured goods from China and Germany. That’s how markets work: goods and money become interchangeable, and if you have money then you don’t need to be able to actually make things any more. Money gives you all the competitive advantage you need.

Except, that’s a strategy which works until it doesn’t. I’m reminded of this bit from Kurt Eichenwald’s piece on Microsoft’s Steve Ballmer in the latest Vanity Fair:

The Microsoft CEO used to proclaim that it would not be first to be cool, but would be first to profit — in other words, i would be the first to make money by selling its own version of new technologies. But that depended on one fact: Microsoft could buy its way into the lead, because it always had so much more cash on hand than any of its competitors.

No more. The advantage that Ballmer relied on for so long is now nonexistent. Google has almost the same amount of cash on its books as Microsoft — $50 billion to Microsoft’s $58 billion. Apple, on the other hand, started the year with about $100 billion. Using superior financial muscle won’t work for Microsoft or Ballmer anymore.

A technology company’s ability to innovate is not a bad metaphor for an economy’s ability to manufacture things and employ people while doing so. If it’s lacking, then for a certain amount of time that hole can be patched with money. But not forever.

And so I think that Barack Obama’s push to bring manufacturing employment back to Michigan and Ohio and Virginia and North Carolina makes all the sense in the world. Trade is not a zero-sum game — Kinsley is right about that — but at the same time that’s no reason to feel sanguine when you see good working-class jobs get exported to countries where the idea of building a blue-collar career in the manufacturing sector is still a perfectly sensible and reasonable one. America does not have a jobs crisis among college graduates, even if the employment situation for recent graduates right now is grim. It does have a jobs crisis in areas where factories have closed and industrial skills are no longer valued.

If you run a company, one of your jobs is to ensure that your company’s money isn’t wasted. And similarly, if you run an economy, one of your jobs is to ensure that your country’s labor force isn’t wasted. There are far too many Americans, right now, who could be working and aren’t. That’s downright inefficient. Their skills are well suited to the manufacturing industry. And so, if new manufacturing jobs are to be created, somewhere in the world, it makes a huge amount of sense that they be created in places like Michigan and Ohio and Virginia and North Carolina: that’s where the low-hanging fruit lies, in terms of hard-working employees with enormous potential productivity gains.

Kinsley’s right that we want China and Germany to have manufacturing jobs. But here’s the thing: China and Germany have manufacturing jobs. New manufacturing jobs, at least in the short term, should move where there’s both a shortage of such things right now, and a large potential labor force willing to get to get to work tomorrow. Certainly it’s the job of the president to encourage precisely that. And if he succeeds, he will have built a powerful economic engine in a part of America which is not doing well at the moment.

Note that Obama talked about bringing those jobs to America not in any protectionist context, but rather in the context of a series of free trade agreements which, at the margin, make outsourcing easier rather than harder. His goal is not to steal jobs from elsewhere, but rather to make America a place where the infrastructure and workers are so attractive that companies around the world decide to source their manufacturing here. No one cares, any more, about the nationality of the employers in these states: a job is a job, whether your employer is American or Brazilian or Swedish.

The computer I’m writing this on was made by an American company in China, but there’s no particular reason why other items in my household shouldn’t be made by a Chinese company in America. If the US can create the conditions for that to happen — if Chinese companies voluntarily move some of their manufacturing here because that’s how effective the US manufacturing sector has become — then everybody wins. That’s what Obama wants. And wanting that requires no misunderstanding whatsoever of how markets work.

Today’s payrolls report was a relatively dull one, showing the same story we’ve seen for the past couple of months: job growth which is barely positive, and certainly isn’t big enough to bring down the unemployment rate or even keep up with population growth.

Go one level down, and you can see the way that the payrolls number is split, between the private sector and the government. This month, the headline growth of 80,000 jobs represented 84,000 new jobs in the private sector, and 4,000 jobs lost in the public sector. And that story — jobs in the private sector growing, while jobs in the public sector are shrinking — turns out to have been in effect basically as long as the recovery has been in place.

The chart above shows the period of the recovery: the years from 2010 onwards when we stopped losing jobs and started gaining them. Or, at least, when the private sector stopped firing people and started hiring people. The government gets the credit, or the blame, for all that job creation. But really, the only job creation that the government can directly control is its own. And since the beginning of 2010, net government job creation is massively negative: we have 536,000 fewer government jobs today than we did in January 2010. (That red spike you see in May 2010 is the census; all those jobs and then some were lost in the following months.) During the same period, the private sector created 4.3 million jobs.

Those lost government jobs are good ones: well paid, with solid benefits. They’re one of the key areas of middle-class employment. And there’s no sign that they’re coming back.

Meanwhile, if you look at the unemployment data, most of the unemployment rates are unchanged this month, with the headline unemployment rate remaining at a woeful 8.2%. But there’s one datapoint which jumps out: the unemployment rate for African Americans soared to 14.4% in June, from 13.6% in May. That’s a number which won’t be lost on Barack Obama, who knows full well that the government is a hugely important source of employment for African-Americans.

In any event, this chart must surely put the lie to anybody who claims that Obama is on any kind of spending spree. Government employment naturally trends upwards, as the population grows and the act of governing becomes increasingly complex. But since January 2009, when Obama took office, total government employment has plunged by 612,000 people. If those people were earning $50,000 each, on average, then that’s more than $30 billion a year in government wages which have been cut.

There’s lots of talk, these days, about how it’s fiscal policy rather than monetary policy which is needed to create jobs and bring down the unemployment rate. That’s true. If the government spent more on important things like infrastructure, that would help create jobs across the economy. But never mind that: the easiest and simplest way for the government to create jobs is simply for the government to stop firing people. The private sector did that more than two years ago. Why can’t the government follow suit?