Like me, you've probably been hearing for years that small businesses are the engine of job creation in the United States. But that's an outdated view. The number of new startup businesses has declined sharply since the beginning of the recession, while the number of jobs created by startup businesses has been declining for over a decade. As this chart from the BLS shows, the number of jobs created by new businesses peaked in 2000, began declining at the start of the Bush administration, and has been plummeting ever since:

The number of new establishments for the year ending in March 2010 was lower than any other year since the series began....The number of jobs created by establishments less than 1 year old has decreased from 4.1 million in 1994, when this series began, to 2.5 million in 2010. This trend combined with that of fewer new establishments overall indicates that the number of new jobs in each new establishment is declining.

....The number of jobs created from establishment births peaked in the late 1990s and has experienced an overall decline since then. The decrease in birth-related employment during the latest recession is the largest in the history of the series, followed closely by the period of “jobless recovery” after the 2001 recession.

Since the recession began in 2008, the biggest net generator of jobs has been neither small businesses nor large businesses. It's been medium-sized businesses.

New York, New York

It's vacation time again! Soon, anyway. I'll be in New York for a few days at the end of June, and I'm looking for suggestions for things to do. Last time I did this I was accompanied by Marian and some friends who had never been to New York before, so lots of standard tourist stuff was on the agenda. This time I'm on my own, so I'd be interested in ideas that are a little off the beaten path. This worked pretty well last time, so I thought I'd ask readers for suggestions again.

Not that the usual stuff is off limits. I'm definitely going to spend a few hours at MOMA. I haven't been there in ages and my hotel is right nearby. And I've never been to the Bronx Zoo. Is it worth a visit if the weather isn't too bad? Or maybe a Yankees game at the new stadium if I can find someone who wants to go with me. Beyond that, though, I have no plans. What should I do and what should I eat while I'm there?

Nobody's reading the blog today, right? So that makes it a good time to revisit a topic at great length that's politically out of the question and will never happen.

I've always been open to the idea of means testing Medicare, and a few days ago I suggested a different way of doing it: after death instead of before. For each Medicare recipient, keep a running tally of the cost of their care, and when they die deduct the premiums and copays they've been responsible for. What's left over gets taken out of their estate, the same way back taxes would. Rich people would end up paying their entire bill, poor people with no estates would end up paying nothing, and those in the middle would pay a portion that depends on how big their estate is. Will Wilkinson wasn't impressed:

I've got a better idea. Don't give the elderly rich any government money for health care. Let them pay for it, because they're rich! And give other seniors just the assistance they need—no more, no less—to buy a health plan of a certain minimum level of coverage. Now, I know this is a fantastical idea for crazed, science-hating, Rand-thumping Jacobins, amounts to destroying Medicare as we know it, and is good for nothing but losing elections. But for all that it seems at least as practical as picking over dead peoples' estates.

That phrase — "picking over dead peoples' estates" — is, of course, the Achilles' heel of my proposal, since that seems to be the instinctive reaction of just about everyone to the idea of allowing the government first crack at estates. Still, let's put that aside for the moment. Is means testing of living people really as practical as means testing dead people, as Will suggests? I don't think it is.

First off, let's review Welfare Economics 101. The problem with means testing — any means testing — is that it acts like a gigantic tax on earnings. Suppose, for example, that you receive $5,000 from the government if your income is below a certain level. If you start earning more, your benefits go down. Maybe the income threshold is $10,000, and for every $1,000 above that you lose $500 in benefits. Do you see the problem? It's like a 50% tax on everything you earn over $10,000, and that reduces the incentive to work hard and earn more.

This is a well-known problem with all means-tested programs, and there's no ideal solution to it. You just have to muddle through. The Medicare version of this is that means testing would reduce the incentive to work and save while you're young. Why bother if it's just going to get eaten up by Medicare expenses later in life? Why not live for the moment, keep your income below the means-testing threshold, and then take advantage of free Medicare when you're old?

Beyond that, there's the problem of how to means test and what the threshold should be. Will says that we should give people "just the assistance they need," but that's not as easy as it sounds. Should means testing be done on income or wealth? If it's income, then you're giving away benefits to people who might have modest retirement incomes but lots of assets. Why should they be allowed to keep their expensive homes and cars and boats and stock portfolios while Uncle Sam pays for their hip replacement? But if you means test on wealth, then you force people to impoverish themselves before they qualify for care. Do you want to be the one to tell granny that she has to sell her house and all her belongings before she gets a dime from the government? I didn't think so.

Well, how about just limiting means testing to the genuinely rich? If you have a retirement income of $200,000 and $10 million in assets, then you can certainly pay for your own medical care. No argument there. The problem is that the genuinely rich only account for about 2% of the population. Maybe 5% tops. Sure, you can make them pay for their own care, but it's not going to make much of a dent in Medicare spending. So why bother?

So now consider my idea. You can earn and save money in your youth and know that you'll still have it in your old age. You can spend it as you like. We don't need any complicated formulas for figuring out who qualifies for free Medicare and who doesn't. We don't need to impoverish granny and take away her house.

Instead, we just keep track of what you spend and then take it out of your estate when you don't need it anymore. Will people try to hide assets or give them away in order to avoid Uncle Sam's bite? Sure. But think about this for a moment. The average cumulative Medicare bill after you've died will be on the order of $100-200,000. The really rich, who have the means and the legal talent to do fancy estate planning, aren't going to run down their estates below that amount. It's just too piddling, and they want to have at least a few millions unencumbered by legal chicanery throughout their lives. Conversely, the working and middle classes mostly don't have the ability (i.e., money for expensive lawyers and estate planners) to cheat their way out of this. That leaves the upper middle classes, and they'll probably try to evade some of their Medicare expenses. But that's a relatively small number of people — and without minimizing the problem here, it really is possible to regulate a lot of it away. If Medicare had first claim on estates the same way the IRS does, it would mostly get all the money owed to it. Just giving them first claim on homes would go a long way toward keeping things kosher.

This doesn't completely get rid of the Welfare 101 problem, of course. There's still a certain amount of disincentive to earn and work while you're young, knowing that you can't bequeath every last dime of your money to whoever you want to. But the disincentive is a lot less. Your parents love you and all that, but guess what: they mostly love themselves even more. They'll do a lot more to protect their own access to their wealth than they will to protect yours.

To some extent, of course, all I've done is shift the problem: there's now an incentive to spend all your money not during your working years but during retirement. Why not, if it's all just going to Uncle Sam after you die anyway? There's no question this will happen, but my guess is that it will happen less you might think. I don't know if there's any empirical evidence on this score (how would you get it?), but there's a limit to how much people want to spend down their wealth. Mostly they don't want to sell their houses while they're still alive, for example, and if they're the saving types they probably want to keep a certain amount of their savings around no matter what. Besides, if you means test Medicare, this incentive to spend down your savings during retirement exists regardless of whether the bill comes due before or after death.

So, roughly speaking, that's my case. Charging for Medicare expenses after death solves the problem of trying to figure who deserves what and how much you can afford. We just don't bother. We simply tot up the charges and then take it out of your estate. If there's no estate, that probably means you were poor and couldn't have afforded to pay for it in the first place. If there's a big estate, it means you were rich and can pay for 100% of your Medicare costs. And for the middle classes, which are by far the trickiest for any means testing policy, it allows effective means testing that, almost by definition, takes from you only money that you can truly afford to pay.

There is, of course, no reason this has to be a standalone policy. We still need to rein in the growing costs of Medicare no matter what. You might also want some pretty strict rules about what you can do with your money if you're currently in a nursing home being paid for by Medicare or Medicaid. (Though the rules on this are already pretty strict in a lot of states, which really do require you to impoverish yourself before you qualify for aid.)

But still: this would almost certainly raise a huge amount of money. It would raise it not based on what you might use in the future, but on what you've actually used during your life. And it would raise that money from people who don't need it anymore.

Let me repeat that: It would raise the money from people who don't need it anymore. If you want to think of this in ghoulish "picking over dead peoples' estates" terms, you can. But it's not. What it is is charging people for a service based on whether they can afford it; it's allowing them to live their actual lives free of fear and impoverishment; and it's settling an account the same way that anyone else would who has a claim on an estate. Do you think of a supermarket as ghoulish if they insist that granny's estate pay for the grocery bill she ran up during her final year of life?

There might be technical reasons that make this unworkable (though I suspect most of them could be resolved tolerably well), but philosophically I just don't see the objection. It's fair, it's efficient, it raises a lot of revenue, and it lets people live their lives decently for as long as they're alive. What's not to like?

Crime guru James Q. Wilson surveys the evidence for why violent crime rates have dropped so dramatically over the past two decades. The state of the economy, he says, seems to have little to do with it:

One obvious answer is that many more people are in prison than in the past. Experts differ on the size of the effect, but I think that William Spelman and Steven Levitt have it about right in believing that greater incarceration can explain about one-quarter or more of the crime decline.

....There may also be a medical reason for the decline in crime. For decades, doctors have known that children with lots of lead in their blood are much more likely to be aggressive, violent and delinquent. In 1974, the Environmental Protection Agency required oil companies to stop putting lead in gasoline....A 2007 study by the economist Jessica Wolpaw Reyes contended that the reduction in gasoline lead produced more than half of the decline in violent crime during the 1990s in the U.S. and might bring about greater declines in the future.

....Another shift that has probably helped to bring down crime is the decrease in heavy cocaine use in many states....Drug use among blacks has changed even more dramatically than it has among the population as a whole....Among 13,000 people arrested in Manhattan between 1987 and 1997, a disproportionate number of whom were black, those born between 1948 and 1969 were heavily involved with crack cocaine, but those born after 1969 used very little crack and instead smoked marijuana.

So if I can put words into Wilson's mouth, the decline in crime is perhaps one-quarter due to increased incarceration, one-quarter due to reduced cocaine use, and one half due to reductions in blood lead levels in children. Better policing might be part of it too, though the evidence is spotty. Oddly, though, Wilson's own summary is different: "At the deepest level, many of these shifts, taken together, suggest that crime in the United States is falling [...] because of a big improvement in the culture." Aside from the reductions in cocaine and crack use, however, none of this sounds all that cultural to me. It sounds like we cleaned up the environment and built a lot of new prisons. It's hard to see an awful lot of room for cultural explanations here.

My sister headed off to England for a vacation yesterday, and before she left she insisted that I post some extra special catblogging today so that she'd have something good to look at while staving off jet lag on her first evening in town. We always do our best on that front, but it really all depends on the cats, doesn't it?

So how's this? On the left, Inkblot is camped out on our new sofa, which Karen hasn't yet been over to see. So this is her first look at it. Inkblot's bright-eyed expression is due to timing: I took this picture last night just as Marian was shuttling food around for dinnertime, and that perked him up. On the right, Domino is basking (as usual) in the sun this morning. However, this picture's extra specialness was probably dimmed a bit because she kept batting the camera strap around instead of holding still in some kind of extra specially cute pose. Still, I hope this does the job.

Have a good Memorial Day weekend, everyone. There will probably be a light bit of posting this weekend, but normal blogging will resume on Tuesday.

Do you like charts and graphs? If you read this blog, there's a good chance you do. So you might want to check out a new search engine, Zanran, which is specifically designed for finding data and statistics. It's still in beta, and I've only played around with it a little bit, but it seems promising. If you give it a try, let us know in comments what you think.

In an aside to a post about how dismally ignorant Republican lawmakers are about the debt ceiling and what it means, Jon Chait says:

As a journalistic side note, I should point out that Politico has really upped its game and is now producing far more good political coverage than any other outlet, as the large number of my items keying off Politico stories would testify.

You know, that seems to be true. I'm not a religious reader of Politico, and it's not as if they've never run good stuff in the past, but the ratio of solid pieces to idiotic gossipmongering does seem to have gone up lately. Anyone else notice the same thing?

Today's Captain Louis Renault Award goes to the Pakistani military:

Embarrassed by the Osama bin Laden raid and by a series of insurgent attacks on high-security sites, top Pakistani military officials are increasingly concerned that their ranks are penetrated by Islamists who are aiding militants in a campaign against the state....Army chief Gen. Ashfaq Parvez Kayani, who like the government has publicly expressed anger over the secret U.S. raid, was so shaken by the discovery of bin Laden that he told U.S. officials in a recent meeting that his first priority was “bringing our house in order,” according to a senior Pakistani intelligence official.

....It is unclear how authentically committed Kayani and other top military leaders are to cleansing their ranks.

You can count me among those who think it's unclear just how authentically committed the Pakistani military is to rooting out Islamists in its ranks. Or that any of its top officials were genuinely "shaken" by the discovery that Osama bin Laden was hiding in Pakistan. Or that they seriously expect us to believe they plan to do anything at all about this. As long as these guys are helpful — or perceived to be helpful — in Pakistan's neverending struggle against the Great Satan India, they'll have an endless supply of high-level sponsors. When they're no longer helpful, they'll get purged. American desires really have nothing to do with it.

Karl Smith notes a Gallup poll showing that half of all respondents think more than 20% of Americans are gay or lesbian:

What makes this interesting to me is not that people are bad at demographics.

It's that I would assume that people’s immediate experience is influencing their estimate of all of America. Yet, 52% of America can’t be experiencing anything like 1 out of every 5 people I know is gay.

So my guess is that most people don’t really get what these numbers mean in terms of their daily life. Of course, to some extent we already knew that but it always interesting to see it come out in actual data.

Yes, people in general are pretty bad with numbers, but I think the real explanation for this is a lot simpler: gay and lesbian issues have been getting a lot of attention in the news lately, and that naturally makes people think they're more numerous than they really are. And personal experience probably has little to do with it. They themselves might know very few gays, but they just figure that's because all the gay people live in San Francisco or Seattle or New York.

The real number, by the way, is around 3-4%.

During the financial crisis the Fed made hundreds of billions of dollars available to European central banks in order to facilitate payments that needed to be made in U.S. dollars. But Bloomberg’s Bob Ivry reports that there was much more going on: the Fed was actually making direct — and very secret — loans to European banks at interest rates as low as 0.01%.

The $80 billion initiative, called single-tranche open- market operations, or ST OMO, made 28-day loans from March through December 2008, a period in which confidence in global credit markets collapsed after the Sept. 15 bankruptcy of Lehman Brothers Holdings Inc.

....“I wasn’t aware of this program until now,” said U.S. Representative Barney Frank, the Massachusetts Democrat who chaired the House Financial Services Committee in 2008 and co- authored the legislation overhauling financial regulation. The law does require the Fed to release details of any open-market operations undertaken after July 2010, after a two-year lag.

....Credit Suisse’s borrowing peaked at about $45 billion in September 2008....RBS’s use of ST OMO hit about $30 billion in October 2008....Frankfurt-based Deutsche Bank’s use peaked at about $20 billion in October 2008, its chart shows.

This is via Felix Salmon, who comments:

Why did the Fed set up a short-term lending program which seems to have been aimed overwhelmingly at European banks? And how does lending $45 billion to Credit Suisse support the flow of credit to U.S. households, in any but the most circuitous manner? It’s probably not worth asking the Fed these questions. But it does seem that the governments of Switzerland, Germany, France, and the UK should all be sending thank-you letters to 33 Liberty Street if they haven’t already done so: it’s entirely possible that the New York Fed bailed out their banks without those governments even knowing about it. That’s just how generous we are, in this country.