Felix Salmon

Why we shouldn’t subsidize construction

Felix Salmon
May 28, 2010 23:05 UTC

I’ve received some great feedback on my post on subsidizing construction loans, especially from Tom Lindmark, who points out that if you want a short-term jobs boost which doesn’t increase U.S. imports very much, then construction is a great area to subsidize. Lindmark cites Mike Mandel in support of his argument, which is always a good sign, but it’s worth noting that Mike kicks off his post by noting that he’s “never been a big fan of home construction as a driver of economic growth”. It’s a cheap high, and the hangover is always brutal.

Meanwhile, commenter winstongator points me to this article from today’s Raleigh News Observer. He does so because Brad Miller specifically singled out Raleigh as a city where new construction was needed. The first line says it all:

Hue, the multicolor building that is the largest condo project ever attempted in downtown Raleigh, closed its sales office without ever selling a unit.

Of course, one 208-unit condo is not necessarily representative of the new-construction market more generally. But its fate does help explain why lenders might be reluctant to throw good money after bad when it comes to construction loans.

Miller, too, emailed with some good points:

Subprime lending was not driven by home ownership; home ownership was the political excuse for it…

Only about ten percent of subprime loans were for the purchase of first homes. More than 70 percent, as I recall, were refinances… and the vast majority of homeowners who got subprime mortgages qualified for prime mortgages…

Oversupply undoubtedly was a result of the bubble in many markets. Of course home builders built like crazy when they could sell houses for much more than houses cost to build. But I suspect that an oversupply of houses is less of a factor in the collapse than is widely assumed…

When we come out of this we’re going to have to reinvent our housing market, by the way. Have you thought much about that?

This is all true; the peak level of home ownership in this country predates the subprime bubble, and the bursting of that bubble can’t credibly be blamed on too much construction.

So if home construction wasn’t to blame for the bubble and bust, and if it’s good for employment, then shouldn’t we encourage it, at least in the short term?

No: the fact is that we’re already plowing far too much in the way of government resources into the housing market. Virtually all mortgages these days are funded by some arm of the government, be it FHA or Fannie Mae or Freddie Mac, and the Fed of course is doing its job too by keeping interest rates at zero, just because the financial system might suffer a massive solvency panic all over again if interest and default rates were to rise again. The housing sector is keeping the government hostage, and the government should be withdrawing from it as much as possible, rather than feeding it even more than it’s doing already.

The difference between Germany and Spain, when you get down to it, is that Germans work for companies which provide goods and services that the rest of the world wants. In doing so, they make good money, which they save up. That’s how they became rich. The Spanish, by contrast, have massive unemployment, and most of the country’s GDP growth in recent years has come from the construction industry. Their main export is tourism, if that counts as an export, and the main way that Spaniards have become rich in recent years is by sitting back and watching the value of their real estate grow exponentially.

The U.S., going forwards, needs to be less like Spain and more like Germany. So let’s not subsidize housing. That way lies fiscal disaster.


HAS THE AMERICAN DREAM BECOME A NIGHTMARE? For decades, the U.S. government has subsidized homeownership, resulting in real estate speculation and overinvestment and contributing to the global recession. Yet Washington has been adding even more subsidies and deepening the federal commitment to the old housing strategy, making it harder to move to a new one.

1. Mortgage Interest Tax Deductions.

2. Artificially Low Interest Rates and Adjustable Rate Mortgages.

3. Tax Credits for first time home buyers was promoted as free money, and Congress extended the program to repeat buyers. This tax credit was like a drug, and the housing industry became addicted. Extending the credit just worsened the problem.

4. Down Payment Assistance. FHA rules made the down payment issue worse by allowing volume builders to give buyers the required 3% down payment through third-party non-profit corporations such as Nehemiah Corporation in California. Builders gave money to Nehemiah, who then gifted the funds to the buyer for a small fee paid by the builder. It was a way to put more renters into homes and line the pockets of homebuilders.

5. Federal Mortgage Insurance. Banks normally want 20% down but will gladly lend when the government insures loans with just 3% down or less. FHA, VA, Freddie Mac and Fannie Mae now guarantee some 80% of all new mortgages and have insured 96.5% of new mortgages so far this year, putting even more of the burden of risky loans onto taxpayers.

6. Low Down Payments. The USDA now even has a zero-down home loan program, and Texas has begun a down payment assistance program. These were homebuilder initiated proposals that are primarily aimed at generating wealth for builders, realtors, mortgage lenders, and Wall Street. Government officials knew, or should have known, that zero-down loans would put borrowers at risk and taxpayers on the hook. That’s because buyers with little or no skin in the game would be more likely to default on loans and go into foreclosure when inflated home values fall below what is owed, or when property taxes or adjustable interest rates rise, or when employment or medical problems arise.

7. Net Operating Loss Carryback. The extension of this tax provision allowed big builders to refile their tax forms and get over $2.6 billion in rebates from taxpayers, a windfall they’ve been using to buy up land at discounted prices.

Posted by JustLaws | Report as abusive

The Bancrofts’ final WSJ capitulation

Felix Salmon
May 28, 2010 22:19 UTC

Remember the war over whether or not Rupert Murdoch would be able to buy the WSJ from the Bancroft family? Sarah Ellison has written the definitive account of the whole story, and now follows up with a couple of startling quotes from the Bancrofts who opposed the sale:

“I feel blessed it happened when it did,” Christopher Bancroft, a Bancroft heir and former Dow Jones director who campaigned against the deal with Murdoch, told me recently. “I’m glad I didn’t get my way.”…

Even Leslie Hill, a Dow Jones director who opposed the deal so passionately that she resigned her Dow Jones board seat in protest, says she doesn’t regret handing the company over to Murdoch. “Just look at how our family would have fared during the recession,” she told me recently.

It was always clear, of course, that Murdoch was offering the Bancrofts far more than their company was worth. But the family had long said that Dow Jones was not for sale, and before selling out they insisted on putting together safeguards to preserve the WSJ’s independence. Inevitably, those safeguards were ignored by Murdoch, and the paper has swung to the sensationalist right.

But rather than regret the sale to Murdoch, the Bancrofts who opposed the sale now are relieved that they lost. They thought that they were standing up for something more important than money — but now it turns out that money would have turned out to be more important after all. Glad that’s cleared up.


Long before Murdoch, the Bancrofts had handed editorial control to the despicable Robert Bartley, who was proudly to the right of Attila the Hun. The WSJ fight was never about “principles” since the family was disengaged from the paper. It was about the difference between being the owners of a famous property and merely being rich. Rich is better.

Posted by steamboatbill | Report as abusive

Marking bank loans to market

Felix Salmon
May 28, 2010 17:47 UTC

Should banks mark their loans to market? The issue — which flared up briefly at the height of the financial crisis, when everybody was wondering whether many of America’s largest banks were insolvent — is back in the headlines, thanks to FASB’s proposed rule change, which Tracy Alloway calls “mark-to-mayhem”.

Cue the predictable response from the American Bankers’ Association:

If implemented, the proposal would greatly undermine the availability of credit by making it difficult to make many long-term loans, the value of which, even if performing perfectly, would likely be reduced on the day a loan is made.

As a curio, before the financial crisis, banks’ fair value numbers were generally above book value. At that point no one really seemed to care about the discrepency, or mark-to-market, for that matter.

It’s not obvious just how much mayhem the proposed rule change would cause, since Businessweek’s Michael Moore says that “changes in the fair value of loans probably wouldn’t show up in banks’ earnings”. But I’m not completely convinced that this is a good idea.

In general, I’m all in favor of transparency in the reports of public companies in general, and banks in particular. So if banks are forced to reveal the true value of their assets, that’s good. But it’s not good if it just results in effective bank runs, where banks with low-value loans found themselves shut out of the repo markets, for example.

It’s certainly true that when any individual bank — like Goldman Sachs, for example — starts marking its assets to market, that imposes a very useful discipline and can help that bank avoid large losses: when the loans start dropping in value, they get dumped sharpish.

But that’s much easier for Goldman Sachs than it is for commercial banks with enormous long-term loan books and valued relationships with their borrowers. What’s more, it would be disastrous if every bank in America started dumping its loans every time they fell in value — given that they would all be sellers rather than buyers, the value of the loans would plunge enormously overnight, and then they really would all be insolvent.

My feeling is that so long this is just an extra reporting requirement, and it doesn’t show up on the income statement or the balance sheet, we’re probably fine. Banks should certainly be marking their loans to market internally, and if they’re doing that it makes sense to ask them to report those marks to their shareholders on a quarterly basis. But there is a real risk here, if those shareholders start to panic when they see the marks.


@MarkWolfinger: Yes, at least some banks will respond internally to internal valuations.

Felix, part of the trouble here has been where banks attempt to hedge their credit risk in the CDS market, where they are required to mark the CDS to market but not the loan it was supposed to be hedging. There has at various times in the last several years — they seem to change these rules from time to time — been an option to designate certain positions as hedges, in which case they don’t have to be marked to market. This naturally opens up some room for shenanigans, but so does treating economically equivalent (or very similar) positions differently.

There’s a continuum of different degrees of liquidity and transparency in market prices. I know that DE Shaw and Goldman have both owned wind farms at various times in the last five years; those are hard to get market prices for. If you’re hedging them with more liquid instruments, those are easier to get prices for, but the whole point is that the value of those instruments should correlate negatively with the value of the illiquid asset; your net worth is less volatile than would be suggested by marking the liquid assets to market and not the illiquid ones. This leads you, eventually, to more of a mark-to-model, where your model assumes that things that correlate negatively will continue to do so. So the only definite conclusion I can give is that you’re not going to find a system that’s perfect; hopefully you can formulate a system with enough of these problems in mind that it doesn’t make any of them too bad.

Posted by dWj | Report as abusive

The problems with university endowments

Felix Salmon
May 28, 2010 15:55 UTC

If you fancy some iPad reading for the Memorial Day weekend, you could do a lot worse than to download this Tellus paper into GoodReader or similar. It’s titled “Educational Endowments and the Financial Crisis: Social Costs and Systemic Risks in the Shadow Banking System; the lead author, who writes very clearly and readably, is Joshua Humphreys.

I’m still working my way through the whole thing, but my initial impression is very positive. Humphreys points out in great detail, for instance, the downsides associated with giving university endowments charitable status and allowing them to issue tax-free bonds; I don’t know what happened to the tax which was being mooted a couple of years ago, but maybe it’s time to revisit it, especially since opposition to the idea even back then was so weak.

Humphreys also notes that university endowments in many ways exacerbated the financial crisis, as well as doing great harm to their own university budgets and their local economies. Meanwhile, their governing boards tend to be incredibly conflicted, with more than half a dozen trustees on Dartmouth’s board alone having managed investments for the endowment.

Humphreys has his own axe to grind: a long-time advocate of socially responsible investing, he makes the case that “as long-term investors, colleges and universities have an important stake in the sustainability of both the wider financial system and the broader economies in which they participate. Rather than contributing to systemic risk, endowments should therefore embrace their role as nonprofit stewards of sustainability.”

This makes sense to me, especially if university endowments are going to operate under the umbrella of charitable status. But even if they want to continue to chase absolute returns, it’s clear that the endowment model massively overestimated their appetite for illiquid assets. The idea was that because they’re investing with the longest conceivable time horizon, they can put a lot of their money into highly illiquid investments. But then they got bit by the fact that their universities were naturally likely to fall back on endowment monies at precisely the point at which illiquid markets seize up completely. Endowments should be countercyclical buffers, when it comes to universtity finances, not pro-cyclical exacerbators of financial crises.

And they should also be a lot more transparent than they are. Writes Humphreys:

When reported, school-specific data are nonstandardized, inconsistent, incomplete and fragmentary, and scattered across municipal, state, SEC and IRS filings, incommensurable annual reports, and costly proprietary financial databases unavailable to the general public.

There’s no excuse for this. Let’s force endowments to standardize their public reports, and show, rather than tell, just what their highly-paid employees are doing to deserve all their millions of dollars in remuneration. And let’s force them, too, to spend a lot more time concentrating on liquidity risk management, and to cast a skeptical eye on the amount of leverage that these institutions really need. Humphreys finds, for instance, a 2007article by Geraldine Fabrikant about Jack Meyer, containing this astonishing number:

When Mr. Meyer and his team were at Harvard, the endowment was known for making money by betting on small pricing differences between different kinds of securities.

For example, Mr. Meyer and his team might capitalize on the price difference between new Treasury issues and older ones. And to magnify gains, they would leverage those bets as much as 15 to 1.

That sounds very much like LTCM to me, and I think everybody can agree that we don’t want university endowments to be LTCM. And I’m not sure that it’s at all possible, with hindsight, to justify these kind of salaries:


It’s true that if you want massive returns on your endowment, you’re likely to end up paying massive salaries to the people who manage it. But you’re also likely to start spending future endowment gains you don’t yet have, and end up with a billion-dollar hole in the ground. It’s good for universities to be ambitious. But not if their ambitions expand to the point at which they feel the need to start selling off their own donated art just to keep the lights on.


I attended a public university with an endowment of $35m and annual state funding of ~$50m (FY09). Although not a great light of Western civilization, it specializes in producing competent teachers and useful graduates in physical sciences. It has a total enrollment of almost 12,000, coincidentally about the same as Harvard, so it is possible to educate the same number of students for what Harvard paid just to manage its endowment (this ignores facility costs and tuition, but Harvard has much to be ashamed of on both accounts). I agree that no sane argument can be made that this is a charitable educational endeavor.

Posted by ORD | Report as abusive


Felix Salmon
May 28, 2010 04:50 UTC

Ryan Avent sides with me over Matt DeBord on congestion pricing — Sustainable Cities

The townhouse condo apartment bought with Kenneth Starr’s (allegedly) ill-gotten millions — Wellcome Mat

Euroeconomics, explained — Clarke & Dawe

“I have an abiding fear that Felix is right” on online advertising — Kevin Drum

Jack Schofield is leaving the Guardian after 25 years there. I learned what the web was from him — Guardian

Someone’s buying puts on the S&P 500 with a strike at 50. What’s the point? — Kedrosky

Atlantic Media Halts Work on Business Website — Bercovici

Overstock.com’s CEO Unloads Shares While SEC Fiddles — Sam Antar


““I have an abiding fear that Felix is right” on online advertising — Kevin Drum”

I am always amazed that there is any arguement for consumer privacy – I always wonder what I am buying at the supermarket that I want to keep hidden. Frozen food? Cheetos? Abnormal amounts of brocoli?

People give up such privacy for a pittance because it is not worth a pittance to them. At one time people feared being photographed. Now, you can go to numerous websites and see more of people than perhaps you should want to.

Posted by fresnodan | Report as abusive

Is there any reason to subsidize construction loans?

Felix Salmon
May 28, 2010 04:34 UTC

Is Brad Miller — one of the most financially-sophisticated Congressmen in the House — really sponsoring a bill in which Treasury would provide loan guarantees for homebuilders? Calculated Risk responds as one might imagine:

I thought this was from The Onion … unfortunately it is not.

My feelings were very much with CR’s, but I thought it was at least worth asking Miller what exactly was going on. He pointed me to a letter he wrote to Treasury a year ago, and added this:

Yes, there is a huge overhang of existing housing units in many markets. You may recall that I’ve been very critical of Treasury for not doing much more about foreclosures, which is pushing down housing prices and adding to the overhang in many markets. Total household net worth declined by more than the GDP over about an 18 month period. Much (probably most) of that was the result of the decline in home values. I think that the loss of household net worth has been as much of a factor in economic anxiety as unemployment, and the diminished consumer spending that has resulted is making a strong recovery very difficult.

That is not true in every market. I certainly don’t suggest that we build more timeshares in Las Vegas, which is obviously comically overbuilt, but the apartment vacancy rate in Raleigh is three percent. Regulators are pushing lenders, generally smaller banks, to limit their real estate exposure without much regard to the demand in the specific market. So we’ve gone from indiscriminate lending to indiscriminate refusals to lend, or worse, lenders are calling existing performing loans. The housing industry has led us out of recessions in the past, which is obviously not going to happen this time, but the inability of homebuilders to get acquisition, development and construction loans in markets with a demand for new housing is making the recession much worse. About 16 percent of jobs in the recent past have been in the housing industry. Unemployment in the homebuilding industry is now about 25 percent. The continued high unemployment in that industry is a huge drag on any recovery, and it’s pretty tough on the unemployed.

The bill requires that the loans only be in “viable” markets, which means not comically overbuilt markets. Treasury should be able to tell the difference, and banks need to as well, since the guaranty is capped at 80 percent of the loan.

I am not new to this issue. I am attaching a letter I wrote to Treasury more than a year ago.

So in short, I agree with you that building houses for which there is not a demand does not build real economic strength, but I think building houses for which there is a demand does.

So, that’s the rationale, and there are parts of it which make sense. If banks are as dysfunctional and self-defeating in their relations with homebuilders as they are in their relations with homeowners, then I’m perfectly willing to believe that they’re destroying projects which can and should be perfectly profitable for them — and in the process causing unnecessary unemployment.

But the fact is that the history of government stepping in and telling banks what to do is a sorry one, whether it’s done by outright instruction, in which case the banks tend to lose a lot of money, or whether it’s done through loan guarantees, in which case the government tends to lose a lot of money.

I daresay that there are homebuilders out there who deserve loans they can’t get. But the same is true of other businesses too — businesses which create the vibrant sectors of tomorrow’s economy that ideally we really want to encourage, rather than crowd out. Banks aren’t going to lend more if this bill passes: they’re just going to shunt their lending from non-guaranteed sectors to homebuilding. And that can’t be good for the long-term health of the economy.

As for the unemployment problem, there’s no doubt that we want the people who have lost their jobs in the homebuilding sector to find new jobs as quickly as possible. But do we want those new jobs to be in the homebuilding sector? Not really. If we’re going to encourage job creation, let’s try to do it in areas of the economy which will help drive exports rather than imports, and which underpin a genuinely strong economy.


Sales stop for Raleigh condo project

http://www.newsobserver.com/2010/05/28/5 04370/hue-condo-sales-stop-no-units.html

“Hue, the multicolor building that is the largest condo project ever attempted in downtown Raleigh, closed its sales office without ever selling a unit.”

Posted by winstongator | Report as abusive

The operations of Annie Leibovitz’s sophisticated advisor

Felix Salmon
May 27, 2010 21:14 UTC

Remember Art Capital Group’s lawsuit against Annie Leibovitz? It took pains to point out the sophistication of her advisors:

Each of the sophisticated parties were represented by able and sophisticated counsel and financial advisors and the likelihood that certain of the collateral would need to be sold to satisfy Defendants’ obligations under the secured loan agreement again was directly discussed with Ms Leibovitz, Leibovitz’ attorneys and Leibovitz’ financial advisors, Starr & Co.

That was enough for Cityfile to decide that Leibovitz cannot have been duped:

Friends of the photographer suggest that Leibovitz had no idea she was giving up so much when she took out the loan; they also seem to be shifting some of the blame to Ken Starr, the financial adviser who took the photographer on as a client in 2007 and who was also responsible for introducing Leibovitz to Art Capital Group. Pinning the blame on Starr, who boasts an insanely long list of celebrity clients, may be a hard argument to make…

Given the list of people who have entrusted their finances to Starr in the past, Leibovitz’s suggestion she wasn’t adequately briefed on the terms of the loan before signing the papers is a tad suspect. Presumably the accountant has dealt with desperate and slightly clueless celebs in the past. And if her plan to somehow extricate herself from the mess by pinning the blame on Starr, she may have an uphill battle ahead.

Now, however, her defense seems a bit more credible, given that Starr has been arrested and charged with five counts of fraud. The complaint gives a good idea of his sales pitch, and of the sophistication of his clients:


Starr’s clearly the kind of person who pushes “secure investments” which at the same time will grow by five to ten times, to clients who believe him. And he’s also Art Capital’s first line of defense against accusations that they were predatory in their loans to Leibovitz. I’m sure they’re very happy, today, that they’ve exited that particular deal.


Who is Client-6: presumably some sort of celebrity, is married, down on his financial luck, dependent on income from jewelery sales, and went to jail in 2009.

Research: Jacob Arabo(v)

- http://en.wikipedia.org/wiki/Jacob_Arabo

- http://abcnews.go.com/Blotter/Business/c elebrity-financial-adviser-kenneth-ira-s tarr-charged-fraud/story?id=10760954

Posted by SteveHamlin | Report as abusive

News consumers, paywalls, and useless tourists

Felix Salmon
May 27, 2010 14:32 UTC

John Gapper returns to the subject of newspaper paywalls today, saying that in the UK, Rupert Murdoch’s Times will have to appeal to a narrow elite if it is to succeed online — something Murdoch has never been comfortable doing. Going after a mass audience online is hopeless, he says, in the face of much lower-cost competitors like the Huffington Post:

Mr Murdoch’s News Corp estimates that the marginal revenue from an occasional browser is less than one tenth of a penny a year. Even Group M, the media buying agency of WPP, the advertising group, argues in a research note that the bulk of news surfers are “useless tourists” who not only pay nothing but have little advertising potential.

What Gapper is recommending to Murdoch is the have-your-cake-and-eat-it holy grail of publishing online: set up a paywall, get subscription revenue, and then see your advertising revenue rise at the same time, when media buyers start being able to target your specific subscriber base rather than just chasing “useless tourists”.

The logic here has existed in print publications for years: newspapers with a cover price tend to have higher ad rates than free sheets, because their readership is more affluent and is also more likely to actually read the paper (and see its ads).

But essentially what’s happening here is that advertisers are using willingness to pay for a newspaper as a proxy for all manner of other desirable traits in newspaper readers, just because there’s no other way of really knowing who’s reading what.

Online, however, is different. Newsletters still exist — where one or two people put out a specialist product, charge a lot of money for it, and manage to make a living on subscription revenue alone. But when it comes to bigger news organizations, no one has even come close to covering their editorial costs with online subscription revenues. So while paywalls can turn out to be an important part of a publisher’s online strategy, a lot of their value comes not from direct subscription revenues but rather from the fact that they allow advertisers to target a specific group of people.

So far, so print-like. But the fact is that online there are much more useful and granular ways for an advertiser to work out who they’re targeting, beyond just saying “we want people who are willing to pay to read this publication”. Media buyers evolve slowly, and they’re used to that model, so they’ll stick with it to a certain extent as they migrate their budgets online. But eventually they’re going to realize that if they stick with that outlook, they’re going to lose access to a huge number of high-value consumers. Millions of people are willing to pay for a physical object — a newspaper — but are not willing to pay to read that same newspaper online. It doesn’t make sense that those millions of people are hugely desirable readers when they’re reading a physical newspaper, and hugely desirable readers if they pay to read content online, but are just “useless tourists” if they don’t pay to read content online.

So advertisers, looking to reach a large audience online, are going to have to look past the simple question of whether or not people are paying for content. And they’re going to end up with a much more granular and useful way of working out who’s seeing their ads: social media.

The fact is that if I sign in to a free site using my Twitter login, I’m actually more valuable to advertisers than if I paid to enter that site. That’s because the list of people I follow on Twitter says a huge amount about me, and a smart media-buying organization can target ads at me which are much more narrowly focused than if all they knew about me was that I was paying to read the Times.

We’re not quite there yet. But it seems to me that online publications are making a big mistake if they make subscribers go through a dedicated registration and login process, because the demographic information they can get from that will be less useful and less accurate than if they outsource the reader-identification procedure to Twitter or LinkedIn or Facebook. And people will definitely enjoy an automatically personalized reading experience, where they can see what their Facebook friends are reading and what the people they follow on Twitter are reading.

At that point, they’re not “useless tourists” any more: they’re highly valuable and targetable news consumers. And the question of whether or not they’re paying for their news becomes much less important to advertisers. And, therefore, to publishers as well.


I only click on ads by accident.

I guess I’m a useless tourist.

Posted by drewbie | Report as abusive


Felix Salmon
May 27, 2010 05:08 UTC

“There is no economic reason for banks to insist on regular capital repayment” — Josh.sg

“Immigration has increased the average wage of Americans modestly in the short-run, and by more over the long-term” — NDN

I do like a good old-fashioned movie review slam. AO Scott vs SATC2 — NYT

“If we like something then we judge it as less risky, and if we don’t like it we judge it as more risky.” — PsyFi

Euro Crisis Dashboard — Inflexionary


“There is no economic reason for banks to insist on regular capital repayment”

Why is this neccesarily a bad thing. Wouldn’t this be the same as the bank (which is the home owner) renting to the inhabitants who have a non expiratory option to buy the house and presumably pay a bit of a premium over a corresponding renting price.

I know that banks do not have a good track record at renting out homes but I see no reason this couldn’t work.

Posted by KevyD | Report as abusive