[Continued from yesterday's Part 1]. 

Yesterday’s post dismantled a slightly fear-mongering Wall Street Journal suggesting that FHA’s net worth would fall below the statutorily mandated 2%, and if so, that FHA would need a ‘bailout.’  In fact it does seem that FHA’s portfolio is under stress, with delinquencies up from 5.4% to 7.8%, and that this will bring its net worth, last year a little over 3% of portfolio, down to or just under the target.

Steering between the capital-reserve pylons

Even though that under-capitalization will be no more than an accounting entry – unlimited credit does have its little advantages, does it not? – it signals a larger challenge, and one that needs addressing:

Critics have said the FHA, which has never had a chief risk officer, isn’t able to manage such a large portfolio in an unstable market.

With the FHA’s entire balance sheet – loans plus insurance commitments – at risk, the portfolio’s health is of critical interest to the nation.  Against that, FHA’s information systems and its risk-management policies have been archaic.  Add this to the list of major initiatives to be undertaken soonest – but in the meantime, keep lending on the successful programs, and shut down the risky or rotten ones – some of them foisted upon it by Congress.

Policymakers have used the FHA to stabilize the housing market by pushing it to offer credit with far easier terms than that offered by most private lenders. For example, it will back loans with down payments as low as 3.5%.

Last year, the agency ended a program [Nehemiah – Ed.] that allowed sellers to fund down payments.

That is fantastic news; I’m chagrined to have missed it.  As I wrote two years ago, in 3% altruistic, it was little more than a transparent scam: 

No, I’m truly substantial 

 Do the non-profits act as surrogates for the buyer, making sure prices are held down?

Sellers have tried to recover the cost of the fee they pay nonprofits by raising the price of the house an average of 3%, a 2005 study [text version here -- Ed.] by Congress’s nonpartisan Government Accountability Office found [also referenced here -- Ed.].   

So let’s keep score here:

·         Buyers use FHA-insured financing for 97% of the purchase price.

·         Even though there is a 3% down payment, buyers do not pay it themselves.  Independent non-profits give them the money.

·         Non-profits get the money from property sellers.

·         Sellers also pay the non-profits a processing fee.

·         Property sellers raise prices by 3%.

In a cruder world, what these particular non-profits are doing might be called ‘procuring.’  In such a procurement, two entities get screwed:

What happens to those receiving a procurement

1.             The buyers get screwed.

The higher prices contribute to the increase in foreclosures by buyers who have no equity in the homes, the GAO found. 

2.             The Federal government (that’s us taxpayers) gets screwed.

The programs are “a contributing factor of increased risk in our portfolio’ of loans, HUD spokesman Lemar Wooley said in an e-mail.  

More than 100,000 low- and moderate-income consumers bought homes using such programs last year. The percentage of foreclosures on these homes is more than double that on other loans sponsored by the Federal Housing Administration, according to agency audits.

Send post as PDF to

By: David A. Smith

Every lender faces the challenge of setting credit policies: tight but not too tight.

That’s too tight

Too tight and the lender loses market share.  Not tight enough and the additional market share acquired turns into loan losses that sink the lender.

A public-policy lender faces the same dilemma but in reverse: liberal but not too liberal.  Too liberal and the government doesn’t merely mirror the exuberant market, it enables that market’s exuberance.  Not liberal enough and it merely eats market share without advancing affordability and helping deserving families improve their housing.

The Federal Housing Administration, the original government lender and hence the precursor to Fannie Mae and Freddie Mac; as a forerunner, FHA’s government connection was and is more direct.  It’s an arm of HUD, hence its securities carry the full faith and credit of the United States, unlike its GSE heirs Fannie and Freddie, which borrowed money on the capital markets without direct Federal support.

[Uh, until recently.  – Ed.]

“It’s pronounced ‘feefa’ not ‘fiefdom’”

That means that unlike Fannie and Freddie, who are likely to return to a non-guaranteed state some day, the government is always exposed for the entire FHA portfolio.  And FHA has a further tool that Fannie and Freddie do not: in addition to making direct loan, it offers mortgage insurance. 

The FHA insures private lenders against defaults on certain home mortgages, an inducement to make such loans. Insurance from the New Deal-era agency has enabled lending to buyers who can’t make a big down payment or who want to refinance but have little equity.

Most private lenders have sharply curtailed credit to those borrowers.

Any FHA-insured loan carries a lender-protection guarantee – upon the occurrence of any non-cured material default, the lender may sell the loan back to FHA, no questions, asked, for 99 cents on the dollar.

Backed by the US government

With the FHA’s entire balance sheet – loans plus insurance commitments – at risk, the portfolio’s health is of critical interest to the nation, and its capital ratios are closely watched, as reported in the Wall Street Journal:

The Federal Housing Administration, hit by increasing mortgage-related losses, is in danger of seeing its reserves fall below the level demanded by Congress, according to government officials, in a development that could raise concerns about whether the agency needs a taxpayer bailout.

‘Bailout’ here is a misnomer.  Unlike private entities (yes, even the GSEs), where capital must be injected, we taxpayers already have guaranteed all their loans.  ‘Bailing out FHA’ is equivalent to ‘bailing out Treasury.’

Agency officials say if there is a shortfall, they don’t have to do anything except report it to lawmakers.

“Your majesties, your eldest son – the swaying fop behind you – as run up enormous debts.”

They are correct; in point of fact, there is no such thing as the ‘FHA insurance fund,’ it’s just journal entries on the Federal balance sheet.

But some mortgage and housing analysts see trouble ahead. “They’re probably going to need a bailout at some point because they’re making loans in a riskier environment,” says Edward Pinto, a mortgage-industry consultant and former chief credit officer at Fannie Mae.

I think Mr. Pinto is being inadvertently misquoted by context.  He has to be too knowledgeable to have spoken of a bailout except in a shorthand sense.

” … I’ve never seen an entity successfully outrun a situation like this.”

Given that followup comment, I suspect Mr. Pinto means that FHA’s capital ratio will fall below the 2% mandated by statute.

Federal law says the FHA must maintain, after expected losses, reserves equal to at least 2% of the loans insured by the agency.

If not, FHA is expected to curtail too-liberal lending practices, but as for the capital, FHA simply passes the buck to Treasury.

All yours, Uncle Sam!

FHA’s taken those risks because it, Fannie, and Freddie, basically are the US mortgage market right now.  When the other lenders pulled back – of their own volition or because of regulatory order – the three F’s kept the money flowing.

Fannie, Freddie, and little FHA!

In the past two years, the number of loans insured by the FHA has soared and its market share reached 23% in the second quarter, up from 2.7% in 2006, according to Inside Mortgage Finance.

Readers will recall, from my post Who’s next?, that Inside Mortgage Finance does a terrific job of tracking and compiling public-domain statistics and interpreting them perspicaciously.  I trust their figures.

Before the crisis, FHA had drifted almost into irrelevance as foolish private lenders (including Fannie Mae, as we documented presciently in Who’s next?) loosened their standards beyond even those of FHA.  Now FHA’s market share has increased ninefold, and it’s still rising rapidly:

FHA-backed loans outstanding totaled $429 billion in fiscal 2008, a number projected to hit $627 billion this year.

Fannie and Freddie have mortgage backed securities.  FHA has insurance.  Same macro-leverage.

We can extract any equity, no matter how painful

Rising defaults have eaten through the FHA’s cushion. Some 7.8% of FHA loans at the end of the second quarter were 90 days late or more, or in foreclosure, according to the Mortgage Bankers Association, a figure roughly equal to the national average for all loans. That is up from 5.4% a year ago.

An additional 2.4% delinquency on a $627 billion portfolio represents $15 billion more in delinquencies. 

The [FHA capital] ratio last year was around 3%, down from 6.4% in 2007.

An additional 2.4% delinquency is not the same as losses.  Loss = Default Rates x Principal-Writedown Given Default.

Let’s do some simple arithmetic.  How much of a writedown can FHA afford on those delinquencies?  40% losses on 2.4% delinquency represents 0.96% lost portfolio value.  In short, FHA had better recover 60 cents on the dollar.  Or else what?

If its reserves fall short, the agency is obliged to notify Congress, which could spark a commotion over the extent to which the government is funding losses in the housing market.

The Journal is mightily pumping its bellows; ’spark commotion’ is a far cry from ‘go bankrupt.’ 

Keep reading, we’re sure you’ll get fired up

A senior official at HUD, which oversees the FHA, said there is “no risk” that the FHA would require money from Congress if the ratio falls below 2%.

The Journal knows full well that FHA is the Treasury – it will keep going indefinitely, losses be damned, as long as Congress keeps giving its authority.

Some housing analysts have said losses might lead the FHA to pull back lending, which has helped boost flagging housing demand.

That is the more practical risk – political, not economic. 

HUD Secretary Shaun Donovan said in June, “there’s a better than even chance that we will stay above the two percent reserve threshold. That suggests, not just for the 2010 business, but overall for the portfolio, that we’ll more than likely to stay out of a broader need for any taxpayer funding.”

Spoken with Shaun’s usual prudence and restraint.

Just trying to lay out the facts

A review conducted this year by mortgage consultants Ann Schnare and Michael Goldberg found the estimated value of the FHA’s reserve could have a shortfall below the mandated 2% minimum of $3 billion during the fiscal year ending Sept. 30, and $4 billion for the coming year.

This is less critical than it would be for a commercial bank, because with the entire balance sheet guaranteed already – think Fannie/ Freddie post-conservatorship – there’s no need to maintain net worth all in liquid form. 

Some economists say the FHA’s lending has been crucial to preventing a deeper bust in property.  

I’m not an economist, but count me among them.

Thomas Lawler, an independent housing economist, said “the alternative could have been a complete meltdown of housing finance” that would have ultimately led to much larger losses.

Together with the GSEs, Fannie Mae and Freddie Mac, FHA has been the mainstay of US housing finance.  Had they stopped lending, we would have been in the soup.

Critics have said the FHA, which has never had a chief risk officer, isn’t able to manage such a large portfolio in an unstable market.

Better late than never. 

We’re going to slap FHA into shape

[Continued tomorrow in Part 2].

Send post as PDF to

Yesterday, in pondering how to finance group-benefit infrastructure like public toilets, we introduced meta-finance, a means of lending to a contemporaneously formed group of people for their group benefit.  On behalf of Development Innovations Group, and as research for a paper, Meta-finance, prepared for DIG as part of DIG’s Housing Finance for the Poor Initiative with the Gates Foundation, and now posted on AHI’s Web site (link in .pdf), it was to seek discovery of meta-finance that, like a nineteenth century explorer, I ventured off to India to meet Mumbai’s Jockin Arputham and the National Slum Dwellers Federation:

Portrait of the entrepreneur with a cell phone and the gift of gab: Jockin Arputham

and Ahmedabad’s Bijal Bhatt and Mahila Housing SEWA Trust:

Bijal Bhatt at her desk

I’ve previously posted my hypothesis that there ain’t no such thing as free infrastructure, which has a corollary: all public-benefit infrastructure has a non-recoverable capital cost.  If you want infrastructure, you need to fund it from a benefactor: a private individual or charity, an emperor, or the state.  This principle appears to hold since antiquity, as presented in an astonishingly apposite article, Thirst: A Short History of Drinking Water, by Duke environmental law professor Jim Salzman.  In future posts, I intend to say quite a bit about Professor Salzman’s research, which is a gushing fountain of historical experience throughout two millennia of human urban habitation.

Both in Mumbai and in Ahmedabad, contrary to what had been reported at second-hand, the fabled creature of pure meta-finance did not exist. 

When I got to India, the gryphon of meta-finance was sleeping

Neither NSDF nor MHT were operating entirely on their own – they had considerable financial help from their municipalities, which in both cases funded the hard costs of building the community-benefit improvements.  So I formulated the Basic Model, which is a public-private partnership with clearly defined roles. 

It’s only a basic model, but maybe it’ll fly

To put on a production, you need a script, a set of players, and a stage.  In meta-finance’s Basic Model, you need:

All the ecosystem’s a stage, the men and women in it merely players

As described in the DIG report:

The counter at the Dharavi communal toilet, 1 rupee ($0.025) per walk-in

For meta-finance to work, the physical improvement whose cost is beyond the members’ capacity has to be fully created by a single funding source (in this case, government).  Incremental construction and incremental financing work fine for private-benefit goods like houses, because people will protect the work in progress.  For a public-benefit construction, you have to protect it until it’s completed (as every subdivision developer knows), which requires that you cannot afford to have construction stop halfway through.  You need the improvements package, and the government funding to get it fully completed.

Jadibanagar, Ahmedabad: streets paved, street lights, and electricity

Courtesy of Ahmedabad Municipal Corporation and Mahila Housing

Then you further need a sustainable operating model: a mechanism whereby the costs of keeping the facility going are paid by the residents, so the whole neighborhood contributes to its upkeep.  In Ahmedabad’s Jadibanagar slum, the public-space improvements are funded with monthly subscriptions that are collected by the savings co-operative.  In Mumbai’s Dharavi public toilet, all resident households pay a monthly membership fee of roughly Rs 40 per household (or roughly five family uses per rupee), which is a very good deal for them since walk-in use costs Rs 1, five times as much.

In the Dharavi public toilet, walk-in charges are assessed, and maintenance is provided, by the operator family, who mans the desk 24/7 because the household lives upstairs, in a flat built as part of the overall package.

Operator’s flat, built on the roof of the communal toilet

The completed Dharavi public toilet is a self-sustaining venture that provides a group benefit, a family’s employment, and even revenue to the community (with the subscriptions covering the base operating costs, the walk-ins generate positive cash flow).

Once developed, it’s that most valuable of inventions, a machine that will go of itself. 

Whoever patents the working perpetual motion machine is set for life

But it needs to be developed, and for that you need the actors, led by the entrepreneurial community development entity.

Mahila Housing’s Ahmedabad office: seeded by USAID, now sustained by Dell

B.         Actors (government bodies, companies) that can use the business financial model

Infrastructure is enterprise, both in construction and operations.  Enterprise is carried out sometimes by government but most often by NGO’s, whether for-profit or non-profit, who use the financial tools to create individual improvement schemes.  

Most of the academic or research literature regarding housing and infrastructure development focuses on the roles played by programs, or by the governments that create them.  Too little focuses on how those programs (which we might consider the DNA of transactions) are turned into actual working properties – by entities acting as independent entrepreneurial enterprises.

It takes three for the match

Jayshree Vyas, CEO of SEWA Bank and Bijal Bhatt, CEO of Mahila Housing SEWA Trust

The ladies who make it go

Self-Employed Womens’ Association (SEWA) Bank

In some situations the roles can be combined (e.g. sponsors and facilitators combine in the global north as CDC’s or Housing Associations) but that combination is not essential.

Like an arranged marriage, meta-finance projects and schemes require three players: a capital provider (the lender/ investor, often a government offspring), a capital consumer (the sponsor), and a matchmaker (the facilitator, usually a CDC or technical assistance entity). 

Normally the equities are fairly balanced between capital and sponsor

Of these, the facilitator is the most active — I say this from over twenty years’ making a living in precisely this role in the US — the sponsor takes the most ongoing risk, with the most operating control, and the capital provider is the one who has to be persuaded to come to the closing table.

Come to the closing table with me …

What does it take to persuade the capital?  A favorable ecosystemic environment, both economically and politically.

C.         Suitable local conditions (favorable environment)

The local environment is critical because each one is different, and its uniqueness defines the amount of government support or subsidy available, the type of physical improvements, the nature of the group entity to be created or elevated, and who can act as the facilitator, including these elements:
 

Personal in-house toilet, Jadibanagar, retrofitted into existing homes

Put it all together, and you have the condensed definition of the Basic Financing Model:

The more I work in housing, the more I come to appreciate the value of controlling the use of urban land.  Land value is a residual, and urban land is valuable directly in relation to what it can be developed into.  Municipally owned land and its corollary, zoning approval, are particularly useful as policy instruments because they are off-balance sheet assets that can be turned into affordability. 

Really, the assets are totally off the balance sheet

You need both the malt and the brewmaster, and of the two, I have more affection for the brewmaster.

Jockin Arputham, facilitator extraordinaire

Send post as PDF to

If you’re French, you invent the pissoir

That’s a fairly simple problem in municipalities where every home has its own indoor plumbing, because the need for public toilets is minimal.  Then too, if you have lots of public accommodations – offices, restaurants, stores – you can require these businesses to have some level of toilet facilities. 

Lavatory, Indian restaurant, Munich, Germany

Such approaches work fine in affluent, developed municipalities, but how do you tackle public sanitation in informal settlements, spontaneous communities (or slums) where people’s habitation has far outrun municipal infrastructure?

There ought to be a viable business proposition, because you have a community of many people, all of whom will benefit if there’s a working toilet block or complex that residents can access. 

The counter at a Dharavi communal toilet

Such a toilet block can also become a paying proposition, with walk-in customers charged a higher rate than subscribers, who pay a lower monthly fee.

What one rupee buys you: a clean squat, a clean flush, and clean toilet paper

So we can imagine a viable neighborhood-level utility, in the form of a community-subscriber public toilet.  Yet the community toilet will not exist unless somebody pays for its construction and operation.  How do we finance it?  We can finance loans to individuals, even if they have only informal income, because we can establish a payment history and can rely on the individual’s motivation.  So if we add together the paying power of a whole neighborhood of residents, each of whom could pay his or her share of the cost, shouldn’t we be able to finance the group improvement? 

Moving from the individual to the group introduces new risks.  As I put it in a paper, Meta-finance, prepared for DIG as part of DIG’s Housing Finance for the Poor Initiative with the Gates Foundation, and now posted on AHI’s Web site (link in .pdf):

In microfinance and housing microfinance, where the loan goes to individual benefit, so the basic equation is one where the marginal benefit to the consumer roughly equals the annual capital cost of the improvement.  In meta-finance, with group benefit, annual capital cost usually exceeds marginal benefit to the consumer (at least as measured by effective demand). 

Anyone who’s ever shared a taxi and tried to split the costs realizes these dynamics of the marginal consumer. 

Who’s paying the marginal cost of all this luggage?

The person who first flags down the cab is willing to pay 100% of the journey’s cost.  Adding another person going to the same destination doesn’t cost the cab-hailer a cent.  In economic-speak:

Once non-recoverable capital costs are introduced and funded, then each additional user (beyond the additional underwriting) can be added to the system to a decreasing marginal cost. 

Anyone who’s ever shared a taxi and tried to split the costs realizes these dynamics of the marginal consumer.  The person who first flags down the cab is willing to pay 100% of the journey’s cost.  Adding another person going to the same destination doesn’t cost the cab-hailer a cent.  In economic-speak:

That economic reality creates the basic paradox of infrastructure: an improvement that benefits many is paid for by a few – a subset.  Those who do not join the initial subscription can benefit nevertheless once the improvement is built. 

In the cab situation, there are only two of us, and usually we split the cab’s cost.  But when the cab is a bus, and there are already 67 of us on it, a 68th person may jump in just before the doors are closing, and get a free ride.  Further, a whole bunch of people may think the same way, lingering by the doors and jumping in when they’re about to close.

This leads to a set of risks absent from those in individual-benefit lending:

Free riders.  Those who did nothing to create or support the infrastructure gain its benefits without paying (e.g. unmetered water from a communal standpipe).

When you’re a kid, you can bum a lot of free rides!

Resource capture and rent-seeking.  Someone in the community seizes operating control over the infrastructure and charges neighborhood residents a premium (relative to the original underwriting projections) that extracts all the affordability for personal (private) benefit.

Toddler in the swimming pool.  The community benefit is not maintained because no one regards it as his or her or their personal responsibility to maintain it (e.g. failure to maintain cleanliness in a communal latrine).

Anybody watching me?

Tragedy of the commons.  The communal good is overused and laid to waste (e.g. latrine whose capacity is exceeded).

Though they’re given economists’ names, these risks are not esoteric; they come up over and over again in any group-use situation.

For meta-finance to be said to exist, these risks must be handled in the transaction structure without reducing it to a series of individual microfinance loans.

Last July, at the Bellagio Urban Summit convened by the Rockefeller Foundation, our Week 1 group (Housing, Water, and Sanitation) tackled this problem – or at least, banged our heads against it.  Led by Franck Daphnis of Development Innovations Group and Melanie Walker of Gates Foundation, we agreed on a group definition of meta-finance:

Metafinance pools individual cash flows to secure previously inaccessible high value loans that provide community rather than individual benefit. Metafinance loans are larger and of longer tenor than traditional microfinance group loans. Items of communal benefit such as potable water pipes, wells, latrines, street paving, drainage ditches, and electrical connections would be ideal candidates for metafinance.

(The definition is from a forthcoming paper on the topic, to be authored by Daphnis and Walker.)

The idea was interesting enough that DIG commissioned AHI to spend a week in India (Mumbai and Ahmedabad) taking a look at two examples of meta-finance pioneers: National Slum Dwellers Federation in Mumbai, and Mahila Housing SEWA Trust in Ahmedabad.  So off I went for a week of looking at Indian communal toilets, and the financing thereof. 

A backpack, a laptop, a universal plug, and a mug of tea: I’m ready!

As described in my DIG report:

How then does meta-finance work – when it works?

The ‘pure form’ of meta-finance is lending not to an entity (even a group-controlled one) but rather lending to a disparate group for group benefit. 

Nearly all real lending is between two parties, a borrower and a lender, where each of them is a single entity.  The lender may be a bank, a credit union, a specialist originator, or an investor, but it’s an entity, with basically one goal: to get repaid and earn yield from the repayments.  The borrower is likewise an entity: a person, a family, a company.  When people want to do something together, they form a new entity – a corporation, a partnership, a co-op, a trust – and that entity becomes the borrower.  The significance is in the binding.  Creating an entity means I and my assets are at risk from your performance.  As a certain group of gentlemen put it, on a momentous occasion:

And for the support of this declaration,

with a firm reliance on the protection of Divine Providence,

we mutually pledge to each other

our lives, our fortunes and our sacred honor.

With this in mind, last October I went off to India to see what was being done and whether it qualified as pure meta-finance, something pretending to be meta-finance, or a partial or emergent form of meta-finance:

Like Eophippus, we all evolve upward from something

Here’s what I found:

In India the pure form [of meta-finance] does not exist, because the lending is done to an entity whose formation is a laborious precondition of any lending.  To develop such group-benefit entities (like co-ops), capital providers to very poor and informal workers, mainly microfinance institutions (MFIs) or co-operative banks), are experimenting with products that touch on meta-finance (because they induce the development of a group-benefit entity as a financial counterparty). 

The Indians I met were finding that, to finance group-benefit improvements, they had first to form the group into a recognizable entity, and establish its membership history and savings record. 

The chair of the Jadibanagar savings cooperative, and her husband

So the idea of meta-finance was forcing a self-organization: the money imperatives were changing the interpersonal relationships, and driving them toward formalization.

That’s not all it takes to finance group-benefit infrastructure, for there are other economic and financial challenges ahead:

The course of true finance ne’er runs smooth

[Continued tomorrow in Part 2.]

Send post as PDF to