GSE regulation: read the menu!
No sooner had the credit-taking begun for the House’s 65-5 report-out of a GSE regulatory bill (to hear the Washington trade associations tell out, each of them single-handedly changed the course of history) than it was matched by the virtually parallel carping about exactly the same legislation:
- The Wall Street Journal, never one to miss a chance to bash Fannie or Freddie, complained that the legislation fell short of what the White House wanted (I missed that day in Civics 101 where they explained how the legislature is merely the administration’s cabinet).
- Affordable housing advocates protested the ‘giveaway’ of allowing the GSEs to increase the ‘conforming loan limits’ (basically, those that count against affordability goals) substantially.
It’s therefore evident that the 65-5 acclamation is illusory, achieved only by giving absolutely everybody (including the GSEs’ friends) everything they wanted. Thus, in time-honored tradition, Congress has merely postponed the tough choices to a later round, with energized constituency groups already trying to get their fingers on the legislative Ouija board to push the planchette their way.
In its tempo, the legislative cycle resembles nothing so much as a grandmaster chess match, with opening moves coming very slowly, reflectively, the subtle maneuvering for position; then, as time dwindles, the moves come faster and the time control (or fiscal year!) is achieved only by a furious banging down of pieces (legislative provisions).
Shall we discuss the budget reconciliation process?
Since we’re in for several months of maneuvering for positional advantage, you may as well settle in for a brief [That's a defined term -- Ed.] introduction into what’s at stake and at issue in the GSE regulation.
Get yourself a cup of coffee, check my GSE archive if necessary, and click Continue. (Or print out the whole piece, in .pdf.)
What’s at stake? The regulatory options menu
Accepting for the moment that the GSEs need significantly improved regulation, the questions become:
A. What should Congress regulate?
The choices (not mutually exclusive at all!) include:
- Safety and soundness
- Business areas and business role
- Affordability commitments
B. Who should do the regulating?
The choices include:
- Specify in statute
- Create a new regulator (independent or affiliated?)
- Parcel out the responsibilities among available regulatory candidates
- Let the GSEs and the capital markets do it themselves (tee hee)
And recall that the choice of regulator can vary by issue. It is entirely possible to have a mixture of statutory, new-regulator, and parceled responsibilities.
1. Safety and soundness: protecting the taxpayers
The GSEs have unique awfully big advantages; they can place US taxpayers at very substantial systemic risk by exploiting their open-ended ‘moral credit enhancement’ associated with the never-stated-but-always-implied expectation that they will be rescued by Congress. There is thus no question that the GSEs need strong regulation to control their risk profile — especially since in the last decade, the GSEs have grown their balance sheets (tenfold or so) and in so doing (a) made tons of money for their shareholders, and (b) put US taxpayers at significant prepayment and interest rate risk by mismatching maturities (lend long, borrow short, and if interest rates rise in the interim, you’re screwed).
Even though the GSEs continue to protest that their magnified balance sheets are helping taxpayers, too many commentators, from Alan Greenspan on down, have demonstrated by logic and GSE history that the GSEs essential function — providing liquidity in the secondary mortgage market by buying loans that others originate, and securitizing them in large pools — can be accomplished without swelling the balance sheet, simply by matching the maturities of the loans bought and the GSE securities issued. So everyone in a position to write law accepts that the GSEs’ balance sheets should be smaller than they are now, and should be contained or managed going forward. How are the balance sheets capped? There are two competing theories:
- Statutory cap. Just as the GSEs have caps on the loans they can buy, it is entirely within the realm of drafting to write a rule that would cap the GSEs’ balance sheets at some multiple of core capital, or of long-term debt outstanding. A statutory cap would be an absolute standard, one that Congress could revisit from time to time (as, for example, Congress periodically revisits the Treasury debt ceiling).
- Strong independent regulator. Alternatively, regulation could be done, as bank regulation is now done, via a strong independent entity that maintains staff capacity and expertise. Indeed, much in HR 1461 is focused on making sure the entity is indeed strong, independent, adequately capitalized, and therefore adequately staff. A regulator with a single mission focus can act quickly and responsively.
A case can be made for either posture. A statutory cap is a permanent shackle — something useful if one believes the GSEs are merely weakened only temporarily, and one mistrusts their motivation once they regain their strength.
But that strength is likewise a potential weakness; if it proved too constricting, it could be changed only by act of Congress — i.e., slowly and unreliably — so the adjustments might be too slow to be of any practical good if the financial markets were moving quickly. An independent regulator could act more briskly, and with a consistency of vision and without the shaky penmanship of legislation (think what the Fed does with interest rates). But independent regulators, especially if they are smaller or less profitable than those whom they regulate, are vulnerable to ‘regulatory capture’ (having their minds clouded by their subject, and succumbing to economic Stockholm Syndrome).
Who knows what evil lurks in the minds of GSEs?
As of now, the Administration has been aggressively touting the merits of a statutory cap, whereas Congress is more inclined to a strong independent regulator.
2. Business areas
By entering a new business area, a GSE does three things:
- Advances the thinking and innovation in affordable housing.
- Expands its franchise and potentially increases the use of its awfully big advantages.
- Possibly intrudes upon or competes unfairly with private capital.
The first is an unalloyed good, the second is a risk, and the third is a policy question. Here there is something approximating a rough consensus: the GSEs’ current ability to move into any business area on their own initiative, while laudable for affordable housing, is too loose a standard. It seems likely that new business regulation, whether via permission-before-action or notice-subject-to-stop-order, will be a function of the new independent regulator that emerges.
3. Affordability
Congress created the GSEs, and endowed them with rights and proprietary advantages under their charters, to further the cause of housing affordability. By definition, therefore, their existence as government-sponsored enterprises is predicated on their using their resources to generate the maximum amount of housing affordability consistent with good business practice.
(Any time that is a bad bargain, the GSEs are presumably free to renounce their charters. Always bear that in mind.)
We can break the affordability mandate down into two basic ideas: (a) affordable housing within core business areas, and (b) affordable housing participating in net profits, through a trust fund or similar approach.
3A. What does ‘affordable housing’ mean to a GSE?
We know that housing demand is elastic — how much housing you consume correlates very well to how much you can afford to pay.
We further know that, since conventional markets find their equilibrium and land value is a residual, over time medium-income households will be able to afford market-quality housing. The logical definition of affordable housing, at least for purposes of measuring GSE performance, should therefore be:
Housing for people with incomes below what the conventional markets reach
In the
I put “
Similarly, in
The gap, by the way, is a function of the set of government incentives for homeownership and rental, including depreciation and capital gains treatment of income-producing property. That’s a subject for a future blog post.
We can also observe that below 30% of area median income (AMI), the housing people can afford is only that which has no economic value. That is, property affordable to people who earn 30% of AMI generates zero Net Operating Income. People below 30% of AMI thus have fundamentally not a housing-supply problem but a poverty problem.
This leaves the realm of 30% to 60% of AMI to be serviced by capital-subsidy programs: soft debt, soft equity, and grant.
Summarizing the income distribution as we find it in the
|
Income level |
Housing consumption |
Policy implication |
|
100% or above |
Homeownership |
Functioning mortgage markets Functioning home resale markets |
|
80% to 100% |
Homeownership, lower end |
Mortgage interest deduction First-time homebuyer programs |
|
60% to 80% |
Rental, conventional market |
Accelerated depreciation |
|
30% to 60% |
Rental, below market |
Capital-subsidy programs |
|
0% to 30% |
Requires income supplement |
The core use of Section 8 |
With modest variations based on regional economic cycles, these ratios have held for over fifteen years. We can thus expect them to persist absent major structural changes in the
3B. GSE affordable lending, goals, targeting and the like
1. Should we have affordability subgoals?
A glance at the preceding chart reveals that affordability levels are tenure-dependent. In other words:
- Market homeownership begins at 80%.
- Market rental begins at 60%.
That leads to several conclusions:
1. Loan activity above market levels should gain no credit. GSE activity on homeownership above 80% AMI should gain no credit. Rental activity above 60% AMI should gain no credit.
2. GSE affordability should be calibrated to sub-goals tied to particular tenure targets. Indeed, HR 1461 (and, for that matter, existing regulatory oversight) establishes one target for single-family, another for multifamily.
3. Subgoals should also target under-served markets such as:
- Rural areas.
- Small properties (e.g. 5 to 50 apartments).
4. Regulation should be dynamic and pro-active. Though equilibrium levels have remained more or less stable over the last fifteen years, that conceals substantial variation both in time and by region. The subgoals will need periodic, probably annual, readjustment (just as CRA standards are calibrated from time to time and market to time). This dynamic tension can be maintained only by a capable independent regulator. To give that regulator strong cards, and to insulate the regulator from later accusations of oversight creep, the specific categories of subgoals should in fact be listed, “including without limitation,” in the reform legislation.
2. What type of loan activity qualifies?
The public-policy goal justifying GSE establishment and existence is the facilitation of new activity, not simply the churning of old activity. But this is harder to specify than it seems at first blush.
The GSEs’ statutory purpose is the purchase of loans that other people originate. GSEs buy loans in packages and raise funds to buy them by selling new securities or loan packages. So if a GSE buys a bundle of loans, are they ‘new’ or ‘old’? Distinguishing secondary purchases that further origination from those that are mere churn is no trivial task.
To take but one example:
The bill provides “full credit toward … the multifamily … goal to dwelling units … that … are financed by tax-exempt or taxable bonds issued by a State or local housing finance agency,”
but only if the bonds are:
“secured by a guaranteed of the enterprise [or] are not investment grade and are purchased by the enterprise.”
So the GSEs can pursue their goal simply by buying tax-exempt bonds issued by HFAs, a community that has historically shown no difficulty whatsoever in raising capital.
This language illustrates the challenges of specifying in legislation; it both prescribes a direction but also prescribes a loophole (or what some might see as a loophole).
Regardless of how much loan activity is specified in the legislation, it should contain a catchall allowing the regulator to disqualify any particular instruments as mere paper shuffling. Probably, simple equity would demand that this disqualification, if it occurred, must occur promptly, and preferably prospectively.
3C. An equity contribution: half-tithing?
As we mentioned, a late amendment sponsored by Mr. Frank requires the GSEs to contribute a half-tithe (5%) of their after-tax income into an affordable housing fund for grants to support affordable housing for the very low income (below 50% of AMI). This is shrewd in several ways:
- It’s tied to profits, which the GSEs are motivated to earn. So though it could be zero in a particular year, all the GSE incentives will work to make it large.
- Purposes are (1) increasing homeownership among very-low-income households (tough!) and (2) increase and preserve the supply of rental and owner-occupied housing for the same households.
- Funds would be used for grants, the softest form of equity imaginable. This will translate into real affordability.
- The deep targeting (below 50% of AMI) is below any market threshold, so the impact will be meaningful.
- It has tight restrictions on the use of funds, and proposes an Affordable Housing Advisory Board to oversee the grants.
The proposal also has other features (one-eighth goes to ‘leveraged grants’ that stimulate economic and community development as well as housing; one-tenth or more goes to homeownership) that seem sensible and will likely attract attention and change between now and legislative enactment.
As the half-tithe trust fund moves forward, keep your eye on:
- How much it is. That’s the easy one!
- When there are exceptions to funding requirements. Too many outs and it will be toothless.
- What it’s targeted for. Aim for continuing frontier-plus affordability.
- Who oversees use of the money. That will be a power nexus of its own.
Finally, bear in mind that legislation has terrible penmanship. The most important goal is a good and enacted bill. It would be a pity if the ice cream cone melted while the children fought over it.
