The Post ran a piece today discussing the agenda of Julian Castro, the new secretary of the Department of Housing and Urban Development Secretary. At one point the piece discusses affordable housing. It then refers to the Johnson-Crapo bill for privatizing Fannie Mae and Freddie Mac. This bill has a provision for a fund that would support affordable housing.
It would have been worth noting the size of the fund. It would get its revenue from a 0.1 percent tax on mortgages issued through the system. If an average of $1.5 trillion a year in mortgages are issued, this tax would raise $1.5 billion annually.
If it costs $150,000 to build an average unit of affordable housing, this fund will be able to support construction of roughly 10,000 units a year, an amount equal to roughly 0.007 percent of the housing stock. Alternatively, if this money was used to subsidize rent, it would provide a subsidy of $1,500 a year ($125 a month) to 1 million households.
Both of these routes may be very helpful to the people who benefit, but they are not of a scale necessary to ensure affordable housing to low and moderate income families. It is worth noting in this respect that there is no dispute that the Johnson-Crapo bill proposal would raise the cost of mortgages. The range of estimates are in the neighborhood of 0.5 percentage points to over 2.0 percentage points.
If we assume that the actual impact is close to a 0.5 percentage point increase, this would imply that a family with a $200,000 mortgage would pay an extra $1,000 a year in interest due to Johnson-Crapo. This is likely to have far more impact in making housing less affordable than the subsidies funded through the bill's tax to promote affordable housing.
One of the factors that made it easy for the housing bubble to be inflated to ever more dangerous levels was the conduct of the credit rating agencies. They gave every subprime mortgage backed security (MBS) in sight top investment grade ratings. This made it easy for Citigroup, Goldman Sachs and the rest to sell their junk bonds all over the world.
There was a simple reason the credit rating agencies rated subprime MBS as AAA: money. The banks issuing the MBS pay the rating agency. If the big three rating agencies (Moody's, Standard and Poor's, and Fitch) wanted more business, they knew they had to give favorable ratings. The banks weren't paying for an honest assessment, they were paying for an investment grade rating.
There is a simple way around this conflict of interest. Have a neutral party select the rating agency. The issuer would still pay for the review, but would have no voice in selecting who got the job.
Senator Al Franken proposed an amendment to Dodd-Frank that would have gone exactly this route. (I worked with his staff on the amendment.) The amendment would have had the Securities and Exchange Commission pick the rating agency. This common sense proposal passed the Senate overwhelmingly with bi-partisan support.
Naturally something this simple and easy couldn't be allowed to pass into law. The amendment was taken out in conference committee and replaced with a requirement for the SEC to study the issue. After being inundated with comments from the industry, the SEC said Franken's proposal would not work because it wouldn't be able to do a good job assigning rating agencies. They might assign a rating agency that wasn't competent to rate an issue. (Think about that one for a moment. What would it mean about the structure of an MBS if professional analysts at Moody's or one of the other agencies didn't understand it?)
Anyhow, as is generally the case in Washington, the industry got its way so the cesspool was left in place. Timothy Geithner apparently is proud of the role he played in protecting the rating agencies since he touted this issue in his autobiography. Geithner is of course making lots of money now as a top figure at the private equity company Warburg Pincus, so everybody is happy.
This is all relevant now because it seems that the rating agencies are back to their old tricks, or so Matt O'Brien tells us in Wonkblog. There has been a flood of new bonds backed by subprime car loans. Apparently Fitch is getting almost none of this rating business because it refuses to rate garbage as AAA.
O'Brien does a good job in calling attention to what is going on in this market, but it would be good to remind everyone of why it is still going on. We do know how to fix the problem. It's just that Timothy Geithner and his friends don't want the problem fixed.
Emily Badger in Wonkblog had an interesting discussion of the issues around state tax incentives to lure or keep businesses. The piece notes that many economists believe that it would be good to ban these incentives since it ends up being a zero sum game. It then includes many comments implying that any bans would be difficult to enforce.
While it is certainly true that enforcement would be difficult, it is worth noting that parallel issues arise in international trade all the time. A major goal of many trade deals is to prevent countries from subsidizing their own industries to give them an advantage in international competition. There are often major disputes over what constitutes a subsidy. For example, Boeing and Airbus frequently end up in suits before the WTO over allegations of unfair subsidies. Nonetheless, few people dispute the desirability of trade agreements attempt to restrict subsidies.
The situation at the state level is comparable. There will always be grey areas as states try to push the limits of acceptable subsidies, but that doesn't mean it is not desirable to outlaw the general practice. Just as with international trade, such an agreement can be expected to substantially reduce the amount of money committed to firm specific subsidies.
The NYT engaged in some mind reading on Gina Raimondo, the Democratic nominee for governor of Rhode Island. In reference to Raimondo it told readers:
"Growing up in a Democratic household, she believed in activist government. (Her father had gone to college on the G.I. Bill.) She also thought pension benefits needed to be curbed to save other government services, not to mention the pension system itself."
It's great that the NYT is able to tell us what Raimondo actually believes about activist government and cutting pension benefits. Most newspapers would just have to report what Raimondo said about her views.
As long as the NYT was doing mind reading it might have been helpful if it told readers whether Raimondo thinks that Rhode Island can break contracts with anyone or whether she only thinks the state has the right to break contracts with its workers. It could also have told readers whether she believes the state has the obligation to respect the law in other areas.
For example, if she wants to provide government services but doesn't want to raise the taxes to pay for them, does she think the state should just seize property to cover the cost, and if so, whose property?
Instead of spending so much effort on mind reading, it might have been more useful to readers if the paper had spent more time examining the specifics of Raimondo's pension proposal. In addition to taking back part of the money the state had committed to pay workers, Raimondo's pension plan also will mean giving hundreds of millions of dollars in fees to Wall Street hedge funds. These fees could easily reduce the pension fund's return by more than a full percentage point.
The Washington Post's Wonkblog had an interesting piece on efforts by San Francisco and other cities to set up rules for short-term rental services like Airbnb. At one point it tells readers:
"critics of any new regulation will likely argue that it imposes onerous bureaucracy on would-be hosts, while setting up a complex system that the city can't maintain."
Actually, it should be fairly easy to enforce regulations by simply holding Airbnb responsible for people who rent through its service. This would leave the enforcement problem with Airbnb. If Airbnb lacks the competence to ensure that its rental units comply with the law, then it will replaced by a more competent business. That is the way markets are supposed to work.
At its peak in 2006, the housing bubble had caused nationwide house prices to rise more than 70 percent above their trend level. This run-up occurred in spite of the fact that rents had not outpaced inflation and there was a record nationwide vacancy rate.
The dangers of the bubble also should have been clear. Residential construction peaked at almost 6.5 percent of GDP compared to long period average of close to 4.0 percent. The housing wealth effect had led to a consumption boom that pushed the saving rate to near zero.
Also, the flood of dubious loans was hardly a secret. The National Association of Realtors reported that nearly half of first-time homebuyers had put down zero or less on their homes in 2005. The spread of NINJA (no income, no job, and no assets) loans was a common joke in the industry.
These points are worth noting in reference to an article discussing the Fed's efforts to increase its ability to detect dangerous asset bubbles. An asset that actually poses a major threat to the economy is not hard to find. It kind of stands out, sort of like an invasion by a foreign army. The failure of the Fed to recognize the housing bubble and the dangers it posed was due to an extraordinary level of incompetence, not the inherent difficulty of the mission.
A NYT article reporting on the economic and political situation in Michigan noted that in spite of the improvement in its economy since the recession, manufacturing employment is still far below prior peaks. It told readers:
"Manufacturing has come back, with payrolls rising to 567,900 this June from 440,600 in June 2009, bringing manufacturing payrolls back to July 2008 levels, but short of the peak of 906,900 in September 1999."
Actually Michigan's experience is not very different from the situation for the country as a whole. Manufacturing employment hit 17,640,000 in 1998. In the most recent data it was at 12,160,000 a drop of 31.2 percent. The 37.4 percent drop in Michigan is obviously larger, but not qualitatively different. The drop did matter more for Michigan because manufacturing was a larger share of employment in Michigan than in the nation as a whole.
The Wall Street Journal devoted a major article to the efforts by President Obama and several governors to address the skills gap. According to the piece, employers in manufacturing can't hire workers with the right skills. If employers can't get enough workers then we would expect to see wages rising in manufacturing.
They aren't. Over the last year the average hourly wage rose by just 2.1 percent, only a little higher than the inflation rate and slightly less than the average for all workers. This follows several years where wages in manufacturing rose less than the economy-wide average.
Change in Average Hourly Wage in Manufacturing Over Prior 12 Months
Source: Bureau of Labor Statistics.
There are workers who have the skills employers need. They work for their competitors. If an employer wants to hire people she can get them away from competitors by offering a higher wage. It seems that employers in the manufacturing sector may need this simple lesson in market economic to solve their skills shortage problem.
Are you scared? How will we pay for that? This is the context that was missing from the discussion of a bill from Utah Senator Orin Hatch which would encourage state and local governments to replace traditional defined benefit pension plans with cash balance type plans tied to an annuity which would be run by the insurance industry.
The piece told readers:
"For local governments and states, the unfunded liabilities are huge, ranging anywhere from $1.4 trillion to more than $4 trillion, depending on the assumptions plugged in by actuaries."
These shortfalls are calculated over the pension plans' thirty year planning horizon, a period in which the discounted value of GDP will be in the neighborhood of $500 trillion. It is unlikely that many readers have a clear sense of the projected size of the economy over this period, so they have little basis for assessing these projected shortfalls. If they did know the projected size of the economy they may disagree with the characterization of the shortfall as "huge." (The difference between the two numbers is based on whether the pension funds calculate their shortfalls assuming that their assets earn their projected rate of return or whether they calculate their shortfall assuming their assets earn the return available on a completely safe asset like government bonds.)
There are a few other points worth noting about this picture. First, the shortfalls are likely to be considerably less next year. Most pensions calculate their current assets using a five year average. Next year 2014 will replace 2009. Unless the stock market plunges in the last three and a half months of the year, this change will lead to a substantial improvement in the funding situation of most pensions.
The second point is that the averages conceal sharp divergences across funds. Most pension funds are reasonably well-funded, with some having funding ratios of over 100 percent. There are a number of outliers, like Illinois, Ohio, and New Jersey, that have badly underfunded plans. This is not due to their investment patterns, but rather their repeated failure to make required contributions.
Finally, it is worth noting that turning over the pension plan to insurance companies will almost certainly raise the fees collected by the financial industry. This means that the same amount of taxpayer dollars will translate into lower benefits on average for retirees. That's obviously good news for the insurance industry, but bad news for taxpayers and public sector workers.
There is one possible policy justification for throwing this money in the garbage. If an insurance company was an intermediary, it might be more difficult for politicians like New Jersey Governor Chris Christie to avoid making required contributions. As it stands now, the refusal to make these contributions appears to be part of Mr. Christie's political shtick, allowing him to portray himself as a tough guy standing up to the state's workers.
If there was an insurance company acting as an intermediary then perhaps the situation may be clearer to the public. Mr. Christie is simply trying to avoid paying bills that he has accrued, effectively stealing money from the state's workers.
Allan Sloan raises an important point about winners and losers from corporate inversions, the process through which a U.S. company arranges to be taken over by a foreign company to lower its tax bill. He points out that many shareholders will be hit with a large individual tax bill because as an accounting matter they will have sold their stock and thereby realized a capital gain.
This isn't a question of shedding tears for these shareholders, who will mostly be in the top tenth or even the top one percent of the income distribution. The point is that this tax scam is not in their interest. While the company may benefit over time from paying lower corporate taxes, this is unlikely to result in a net gain for those current shareholders who have to pay capital gains taxes because of the inversion.
Sloan points out that the big gainers are the financial firms that arrange the deals, who can count on hundreds of millions in fees from a major deal. The corporate insiders (top management) may also stand to gain since they are unlikely to be faced with the problem of having to pay taxes on large amounts of unrealized capital gains.
If the point is to change practices such as corporate inversions, rather than just complain about them, it is important to recognize these distinctions. The financial sector and the corporate insiders are incredibly powerful interest groups. If some number of wealthy shareholders can be brought into a coalition to restrict this sort of tax gaming, it would have a far greater chance of succeeding. (The same story applies to bloated CEO pay, which most immediately is money out of shareholders' pockets.)
When countries went into recessions in the past they usually came out with a year or two of rapid growth that more than made up the ground lost in the recession and then resumed a normal growth path until the next recession. That hasn't been the case in any major wealthy country following the 2008 downturn, although some countries, notably those in the euro zone, have done markedly worse than others.
Perhaps it is this comparison to the weak performance of the euro zone countries that led a piece in the NYT Dealbook section to tell readers:
"Britons also see a Continent that is plagued by deflation and stagnation while their economy has staged a fiery comeback from the financial crisis."
According to the I.M.F. the U.K. economy will be 1.5 percent larger in 2014 than it was in 2007. This would be equal to a bit more than a half year of growth in normal times.
Source: International Monetary Fund.
It is also worth noting that this piece seems to imply that the loss of part of its financial sector would be a big hit to the U.K. economy. This is not clear. Economists usually assume that economies tend to be at their full employment level of output in the long-run. If this is the case, then the loss of banks in the U.K. and/or Scotland would lead the people currently employed in finance to move to other sectors where their labor could be employed productively.
The Washington Post thinks its fantastic that Rhode Island broke its contract with its workers. It applauded State Treasurer and now Democratic gubernatorial nominee Gina Raimondo for not only cutting pension benefits for new hires and younger workers, but also:
"suspending annual cost-of-living increases for retirees and shifting workers to a hybrid system combining traditional pensions with 401(k)-style accounts."
In other words, Ms. Raimondo pushed legislation that broke the state's contract with its public employees. The Post's argument is that these pensions were expensive and the state couldn't afford them. This is not clear. (The Post again played the Really Big Number game telling readers about the $1 trillion projected shortfall in state pensions. That is a really big number and is supposed to scare readers. If it was interested in informing readers it would have told them the shortfall is equal to about 0.2 percent of projected GDP over the thirty year planning horizon of public pensions.)
Anyhow, if the state of Rhode Island really can't afford to pay its bills, why should public sector workers be the only ones to pay the price. The state has hundreds or even thousands of contractors. Why not short them all 10 or 20 percent of their payments? That would be the fairest way to deal with the situation if the state really can't pay its bills or raise the taxes needed to do so. Obviously the Post doesn't believe that contracts with workers are real contracts.
A NYT editorial on Japan's economy may have created false alarms by noting that its economy shrank at a 7.2 percent annual rate in the second quarter. This is true, but it is important to point out this plunge followed a first quarter in which it grew at a 6.0 percent annual rate. The net for the first two quarters is still negative, and the editorial is correct to raise warnings about the impact of sales tax increases on growth, but the picture is not nearly as dire as the second quarter figure taken in isolation suggests.
While the piece also reasonably calls for Japan to remove obstacles to women working and to advancing in the corporate hierarchy, it is worth noting that the country has already made substantial progress in this area. According to the OECD, the employment rate for prime age women (ages 25-54) is actually somewhat higher in Japan than in the United States, 71.4 percent in Japan compared to 69.9 percent in the United States.
Nope, I'm not kidding. We've seen a sharp slowdown in health care costs across the board over the last seven years. This has led the Congressional Budget Office to lower its deficit projections. In fact, the reductions in projected deficits due to this slowdown has been sharper than the reductions that we might have seen as a result of almost any politically plausible cut in benefits. But Robert Samuelson is not happy. He tells readers:
"No one truly grasps why Medicare spending has slowed so abruptly. A detailed CBO study threw cold water on many plausible explanations. What we don’t understand could easily reverse."
In other words, just because the problem seems to be going away doesn't mean we still shouldn't make cuts to benefits. One factor that may lead us to believe that lower cost growth can be maintained is that the United States still pays more than twice as much per person for its care with nothing to show for it in terms of outcomes. In fact, if our costs were the same as those in any other wealthy country we would be looking at huge budget surpluses, not deficits.
The difference in costs is attributable to the fact that our doctors, drug companies, and other providers get paid twice as much as their counterparts in other wealthy countries. Of course these are all very powerful lobbies so we more often hear about proposals to cut benefits for seniors rather than reduce the money being paid to providers.
As noted before, since Social Security payments come from a designated tax, there is no real way to get money for the rest of Samuelson's agenda unless we tax people for Social Security and then use the money for the military or other purposes. Such a scheme is not likely to be very popular and few politicians are willing to openly advocate it.
The NYT Magazine had a piece asking whether subprime mortgages are coming back. The gist of the argument is taken from an Urban Institute study arguing that if we had the same lending standards in place as in 2001, there would have been 1.2 million more purchase mortgages issued in 2012. It goes on to tell readers that reduced sales are holding back the housing market and the recovery. All of these claims are questionable.
First, asserting that 2001 is an appropriate base of comparison is rather dubious. House prices were already rising considerably faster than the rate of inflation, breaking with their long-term trend. Furthermore, the number of home sales had increased hugely from the mid-1990s, which were also years of relative prosperity. If the Urban Institute study had used the period 1993-1995 as its base, before the beginning of the bubble, it would have found few or no missing mortgages. It is a rather herioic assumption to pick a year of extraordinarily high home sales and treat this as the reference point for future policy.
The idea that we should expect more sales to be a big trigger for the economy is also dubious. The factor that determines building is house prices. House prices are already more than 20 percent above their trend levels. There is no reason that we should expect prices to go still higher or even that they would necessarily stay as high as they are now. One factor that is likely suppressing construction is the fact that vacancy rates are still above normal levels. It is understandable that builders would be reluctant to build new housing in a context where there are still many vacant units available.
Finally, it is worth noting that it is not clear that many people are being harmed by not being able to get a mortgage. As the piece notes, the benefits of homeownership are often overstated. With prices already at historically high levels, homebuyers have a substantial risk of price declines and little reason to expect that the price of their home will rise by more than the rate of inflation.
Furthermore, there are large transactions costs associated with buying and selling a home, with the combined buying and selling costs equal to roughly 10 percent of the purchase price. This means that people who move within 3-4 years of buying a home will likely lose on the deal. This is an especially important point if the marginal homebuyers are younger people who are likely to have less stable family and employment situations.
The latter point is especially important in a context where much of the elite is demanding that the Fed raise interest rates to slow job growth. If the labor market does not improve much further, then it is likely that many people will find themselves in a situation where they have to move to get a job. That is not a good situation for a homeowner to be in.
That's not exactly news, but Neil Irwin does a nice job summarizing the data in the Fed's new Survey of Consumer Finance. The item that many may find surprising is that median wealth was lower in 2013 than it was in 2010 is spite of the boom in the stock market over this period. As Irwin explains, this is due to the fact that most middle income families own little or no stock, even indirectly through mutual funds in retirement accounts.
For people near the middle of the income distribution their wealth is their house. In 2010 house prices were still headed downward. The first-time homebuyers tax credit had temporarily pushed up prices. (The temporary price rise allowed banks and private mortgage pools to have loans paid off through sales or refinancing, almost all of which was done with government guaranteed loans.) After it ended in the spring of 2010, prices resumed their plunge, especially for homes in the bottom segment of the market.
Price began to turn around in 2013, but adjusting for inflation they were still about even with where they were in 2010. In many areas the prices of more moderate priced homes were still well below their 2010 levels. This would explain why wealth for families near the middle of the income distribution would be below its 2010 level.
The NYT tolds readers that Google is the victim of a European backlash against U.S. technological dominance. In addition to anti-trust and privacy issues being raised with regard to Google, the piece also notes that Apple and Amazon are being investigated for their tax practices, taxi drivers have protested against Uber, and Facebook is being investigated for anti-trust violations.
It's not clear that any of this amounts to an anti-American backlash. Apple and Amazon constantly face tax issues in the United States as well. Taxi drivers in the United States have protested Uber. And it would not be surprising if both Facebook and Google face anti-trust issues here as well.
But the last two paragraphs go furthest to undermine the European backlash story:
"Then there is Microsoft, Google’s longtime nemesis, which spends three times as much in Europe on lobbying and similar efforts. ICOMP, a Microsoft-backed group, has long targeted Google.
"'Google is clearly in the cross hairs,' said David Wood, a London-based partner at Gibson, Dunn, one of Microsoft’s law firms, and legal counsel at ICOMP. 'A lot of the aura has faded, and the shine has come off, and people don’t think they’re the good guy anymore.'"
Companies often try to use government regulation to hamper their competitors. It is not clear that anything about the actions against Google reflect the "European backlash" promised in the headline as opposed to the sort of opposition that any large company would likely face regardless of the country in which its headquarters are located.
The proponents of fracking have made many big claims about its economic benefits. In addition to lower cost electricity, we are also supposed to get energy independence and a boom in jobs. The NYT picked up this theme with an article that touted an "energy boom" that is lifting the heartland. The piece claims that fracking related jobs have revitalized Ohio's economy with Youngstown being at the center of the action.
The piece tells readers:
"Here in Ohio, in an arc stretching south from Youngstown past Canton and into the rural parts of the state where much of the natural gas is being drawn from shale deep underground, entire sectors like manufacturing, hotels, real estate and even law are being reshaped. A series of recent economic indicators, including factory hiring, shows momentum building nationally in the manufacturing sector."
"New energy production is 'a real game-changer in terms of the U.S. economy,' said Katy George, who leads the global manufacturing practice at McKinsey & Company, the consulting firm. 'It also creates an opportunity for regions of the country to renew themselves.'"
That sounds really impressive. Unfortunately the data do not seem to agree. The Bureau of Labor Statistics shows that manufacturing employment in Youngstown is still down by more than 12 percent from its pre-recession level as shown in the figure below. There is a comparable story with Canton.
Manufacturing Employment in Youngstown
Source: Bureau of Labor Statistics.
While fracking jobs may have helped bring these areas up from the troughs they experienced at the bottom of the downturn, employment in both metropolitan areas is still far below 2007 levels. No one thought either city was booming at that time. In short, the data do not seem consistent with the story told in this article.
The NYT gave readers only part of the story in an article on the Democratic primary race for governor of Rhode Island. It notes that state Treasurer Gina M. Raimondo is currently the frontrunner.
It then told readers in reference to Raimondo:
"The 'tough choice' was her overhaul of the state’s pension system in 2011. She marshaled the state’s Democratic political establishment to increase the retirement age, cut benefits and suspend annual cost-of-living adjustments for state employees until the finances of the underfunded system improved. The move was meant to save $4 billion over two decades and slow state property tax increases. ...
"The pension overhaul is now at the center of a primary race for governor that has become one of the most divisive in the country."
Raimondo did not just cut benefits. She also invested a large portion of the state pension fund with hedge funds and private equity companies under terms that were not disclosed to the public. (Raimondo formerly worked with a hedge fund.) The state's major newspaper has sued (unsuccessfully) to force disclosure of this information.
However the issue is not just cuts to the benefits promised public sector workers. There is also a question of whether the state's pension funds are being used to enrich Wall Street.
Robert Samuelson apparently believes it would have based on his column today calling for more military spending. There are a few points worth noting about this piece.
First Samuelson compares current spending at 3.4 percent to the post-World War II average of 5.5 percent of GDP. For most of the post-war period we were engaged in a military build-up to counter a rival super-power (the Soviet Union). The average also includes long periods of actual war (Korea, Vietnam, Iraq I and II, and Afghanistan). It should not be surprising that at a time when the country is not nearly as engaged in armed conflicts, and faces no major foe, it would spend less on its military.
Samuelson apparently wants the money for the military to come at least in part from spending on seniors, commenting at the end: "Democrats who will cut almost anything except retirement spending."
The cuts to retirement spending that Samuelson wants are problematic. Social Security taxes are designated for Social Security. Samuelson might not have a problem taxing people for Social Security and then using the money for the military, but the public might have a problem with that idea, as would the people who depend on their votes.
There are substantial potential savings in Medicare, but this is because the United States pays more than twice as much per person for its health care as other wealthy countries. However getting savings would require cutting the incomes of doctors, drug companies, and medical equipment suppliers. These are all very powerful lobbies which Congress is reluctant to challenge. While Samuelson implies that the issue is seniors getting benefits that are too generous, the cost issue to the government is that we pay too much for the same benefits that people get in all wealthy countries.
I'm a big fan of nature and hiking, but that number doesn't sound quite right to me. The Washington Post had an article on the recreation business in which it told readers that the country spends $646 billion a year on outdoor recreation and related spending. This figures comes to a bit more than $2,000 per person. If we assume that half of the public doesn't really do anything that fits the bill, then this means the other half spend $4,000 per person per year on outdoor recreation. That comes to $16,000 per year for a family of four.
Let's see, you can a pretty nice tent for a few hundred dollars, hiking boots can cost $150-$200, a good sleeping bag in the same range. That could get us to $700, but of course you don't buy these things every year. If you assume they last an average of 3-4 years, these items will only get you about $200 per year, less than one twentieth of the way to our $4,000 target.
According to the article, the $646 billion figure came from an industry group. The link does not go to a report that could explain how they got the number, but rather a map showing a state by state breakdown. It's not clear how the industry group came up with its number, but it's virtually certain they included many items that most of us would not consider spending on outdoor recreation. The Post should be a bit more careful in uncritically accepting numbers from industry groups.
Robert Salzberg points me to a link later in the piece that goes to the study itself. The study shows that the bulk of its $646 billion in spending is based on food, entertainment, lodging, and travel related expenses. This presumably means that if someone flies across country to visit family members and also goes to a national park then the study would count the air fare and the money spent on lodging throughout their trip. The study does not describe the methodology in full, but it does give a non-working URL as the location of a technical report.