PolicyCast: Capital Rule Jeopardizes Housing, Taxpayers

Kirk Willison: From Washington DC, it’s the inaugural edition of the Arch of Mortgage Insurance PolicyCast a podcast about pressing public policy issues, shaping housing in America. I’m Kirk Willison, for more than a decade policy makers have struggled to resolve the future of the nation’s housing finance system, principally what to do about Fannie Mae and Freddie Mac, the so-called government sponsored enterprises, which have been in government conservatorship since 2008. And over that same time period, my first guest Jim Parrot has been at the center of the debate. During the Obama administration, Jim was a senior advisor at the National Economic Council where he led the team charged with counseling the cabinet, and the president on housing issues. He was the administration’s point person for developing and articulating positions before Congress, the media, and the public. Now, Jim is a partner in the consulting firm, Parrot Ryan advisors. He also serves as a nonresident fellow for the Urban Institute where along with his business partner, Bob Ryan, and Moody’s Chief Economist Mark Zandi. He has written a new report on the latest proposal from the Federal Housing Finance Agency to recapitalize Fannie and Freddie

Kirk Willison: Jim, welcome to the first ever Arch Mortgage Insurance PolicyCast. Thank you very much for coming here.

Jim Parrot: Oh my pleasure. Glad to help.

Kirk Willison: Before we discuss the FHFA’s new proposal or your new Urban Institute report, Jim, how did you first get involved in housing policy?

Jim Parrot: That’s a good question. So I actually came to housing policy, for the first time back when I joined the Obama administration, as you recall, housing was a pretty interesting policy area back then in the sort of dark days of early 2009, I should say. So when I joined the administration, it just seemed like the most interesting way to help. And so I started out as Secretary Shaun Donovan at HUD’s senior advisor on housing finance issues and was there about a year, before the White House sort of plucked me up and put me over the national economic council to help coordinate housing policy administration wide. And that’s where I was for the remainder of the first term.

Kirk Willison: As I mentioned, in my introduction, the government sponsored enterprises or GSEs have been under conservatorship since 2008, including eight years during Obama Administration. What were your goals during that time period?

Jim Parrot: Yeah. So back then, the focus really for those of us in the administration and it sounds surprising in retrospect now in a way, was more to really replace Fannie and Freddie than it was to reform them in the way that we’re focused on it. Now, there was a pretty widespread consensus among policy makers particularly in Congress that the model that we were operating under that we’d had to put in conservatorship was so fundamentally flawed that it needed to be tossed out altogether and not just fixed, and so in the administration, we spent a lot of time focused on frankly justifying for policy makers why you needed a government backstop at all, because there’s a lot of skepticism, especially among Republicans that  a backstop was helpful. There was a widespread view that it actually distorted the market and had some role in causing the crisis to begin with.

Jim Parrot: So much of the time in the early years was spent pushing back against that narrative in defending the notion that the government was playing a useful role in the system if we just sort of revised the way that role might work. And so as we develop our thinking on it over the first couple of years in that first term, we sort of played with various ways in which the government might play a pretty remote risk remote role. And backstopping the system with a lot of private capital in front of it. But just enough of them that they could be allowed to fail, to get rid of the too big to fail dynamic that we had viewed as problematic in the Fannie and Freddie world. But that’s kinda where our heads were back in 2009, 2010, both the administration and a little more broadly in Congress, too.

Kirk Willison: How has your view of the need for reform changed since then?

Jim Parrot: It changed pretty significantly. It’s much trickier to start from scratch than I think we had understood back in 2009, that the degree to which the mortgage market and, really the broader economy has Fannie and Freddie’s infrastructure really embedded in it. It makes it really hard to replace that infrastructure in a way that isn’t overly risky, overly disruptive. So I mean your odds of replacing something that doesn’t work adequately are pretty significant. And then the cost of failure is dramatic given the role of the GSEs. Then even if you get it right, that is even if you come up with an infrastructure that makes sense, the disruption and getting from here to there is pretty significant. So I think I’ve come away with the notion that rather than starting from scratch which is too risky and probably not worth the effort. It’s better to focus on sort of protecting what works well in the system, and overhauling what doesn’t.

Jim Parrot: The way I think about that is the GSEs infrastructure, what they manage today works quite well, and what doesn’t work so much as having that infrastructure owned and operated by two privately owned too big to fail companies who ended up being too powerful for the system leads too much risk in the system distortions and so forth. So all of that has led me to think that the smarter courses to take that infrastructure that works so well and find a way to migrate it into something that functions a little more like a public utility. So that you maintain what works and you sort of migrate away from the ownership structure that doesn’t work well. There are a whole bunch of different ways to do a utility version of housing finance, but the key is to keep it relatively modest in scope, that is the utility’s role, modest in scope, really the role of intermediaries. So you maximize the space that can still be dominated by competition and the light. But anyway, so it’s a far cry from where I started in 2009.

Kirk Willison: The Trump Administration took office in 2017. Are their goals the same as what the Obama Administration’s were?

Jim Parrot: Oh Lord, no they’ve really turned the idea of GSE reform that I think most of us had had coming into this administration on its head in a way. I think most people on both sides of the aisle and probably even more acutely on the republican side of the aisle had viewed the problem that you’re trying to solve as the dominance of a privately owned duopoly at the center of the system, the idea that is just inherently distorting and problematic. And so all the proposals bouncing around were proposals for fixing that problem, and the administration now has us on a course just to take us back to that world. And so it’s a little head scratching. I wouldn’t have guessed four or five years ago that a Republican administration would be putting a certain course in essence back to a sort of hyper capitalized version of 2006, 2007. But, that’s the course that got us on for reasons that I have a hard time following.

Kirk Willison: This past May the Federal Housing Finance Agency, or FHFA released a new proposal for the capital standards for Fannie Mae and Freddie Mac. It’s actually a revised proposal from one that was issued in 2018. How different are the two, can you walk us through the basic structure of the new proposal?

Jim Parrot: It intends to mirror frankly the Basel framework that banks operate within. So it’s got a risk based capital component that has levels that fluctuate with the risk the GSEs were taking at a given time. And then it’s got this was intended to be sort of a backup risk and variant component, which is a leverage ratio based on their assets. The first is what’s complicated, it’s got two basic components that are kind of a baseline of risk based capital, and it’s got some buffers that sit on top of the baseline. The baseline is in essence a requirement to hold capital equal to 8% of the institution’s risk weighted assets, so they calculate the risk weighted assets on the basis of two methods. One is they just take this grid that FHFA has given them a sort of credit risk grid for lack of a better way to put it some LTV, FICO boxes, and the like and the other way to calculate the method is, and the boxes tell you which risk weight per credit risk bucket to apply the multiply that times 3% and it gives you your overall number.

Jim Parrot: Then the second method is using the GSEs internal modeling, which the FHFA has helped them develop and approve. And when you applied both those models, whichever spits out the biggest number, that’s their sort of baseline risk based capital requirement. Then on top of that if and when the GSEs want to pay out dividends to their investors or pat bonuses to their executives, they’ve got to provide or come up with three extra buffers. One of them is a stress buffer, so basically under a time of stress in which their baseline risk based capital is wiped out, they need an extra layer of capital needed to remain no ongoing concerns. There’s sort of a going concern buffer. The second buffer is sort of like a siffy buffer, and it’s intended to sort of capture the extra amount of risk that their failure would impose on the rest of the system. So it’s like the siffy buffer, something that’s intended to keep them sort of extra safe. And then the third buffer is a countercyclical buffer, which is one of the buffers that’s supposed to be built up and in good times, and then drawn down and in times of stress and those three buffers together when you sort of add them up and you put them on top of the sort of baseline risk space capital buffer, the whole thing adds up to about $234 billion according to FHFAs estimates, almost a hundred billion of that comes from those buffers.

Jim Parrot: So a lot of it comes to the risk and variants parts, and that’s based on, I believe the assets at risk. So the GSEs, around September last year, that’s in one of FHFA phase releases maybe in the proposal itself, and then the backup risk and variant leverage ratio that I mentioned is just 4% of their gap assets. In that same analysis that FHFA came out with Fannie and Freddie would hold 245 billion together to meet that second requirement. So the latter would be higher. And so whichever of the two is higher, that’s what they got a hold. We did some analysis in that paper that you mentioned, and according to our looking back historically at the GSE’s behavior over time, the leverage ratio is likely binding more often than the risk based capital part of it.

Jim Parrot: So that 4% is likely going to be an invariant number if it were applied, one thing to add to the basic structure, though, when trying to get your head around how much they’re going to hold to meet their requirements like banks, like MI’s they’ll almost surely hold a cushion over and above whatever their regulatory requirement is just to take account for fluctuations in markets and their risk fluctuations in their capital and so forth to make sure that they don’t within a quarter you know bounce down below their level. So you sort of do your math and come up with whatever the higher of the two levels is. Then you add that extra cushion on there, and you get a sense of what they’re likely going to hold and meet all of this.

Kirk Willison: Are there any other features of the proposal worth noting?

Jim Parrot: Yeah, there’s a lot to it that’s worth digging into, and what you care about probably depends on where you’re sitting in the market. The first and probably most controversial is their treatment of credit risk transfer. So in the current regime, the GSEs have got a fair amount of relief for their credit risk transfer which makes sense. They’re in transferring credit risks, they’re holding less and they should hold less capital. That’s incentivized the GSEs to do a ton of CRT as it’s called. So buy our estimates they are transferring around two thirds of their risk on the loans they guarantee. So that’s a lot of what they do. For reasons that is a little hard to follow the proposed rule really reduces dramatically the amount of relief you get for CRT by some estimates like by half, which is probably going to put CRT out of the money at many points in the market. They do this sort of in two ways. First, they do it directly just by literally reducing the amount of direct relief the GSEs will get. The second one they do it indirectly with this often binding leverage ratio that is the leverage ratio is so high, and so frequently going to be higher than what the credit risk related risk-based requirements are that when it kicks in as will often be the case they just won’t be any incentive to do CRT’s. Because part of the reason why you do CRT is to reduce your capital burden, and we’ve got the leverage ratio kicking in you can’t reduce your capital burden with it, so the net effect on CRT could be pretty dramatic.

Kirk Willison: Let’s take a deeper dive into one of the subjects that you mentioned, credit risk transfer or CRT. Now I should note that Arch Mortgage Insurance is an active participant in the credit risk transfer market, both for the GSEs and for our own portfolio. What is CRT and what has the GSEs done that helps it reduce exposure to the taxpayers?

Jim Parrot: Yeah, so CRT put simply is the way that the GSEs have been able over the last half decade or so to sell off a really large percentage of their credit risk. As I mentioned, I think they’re offloading as much as two thirds of their credit risk today. And as a result, they reduce the risk to the taxpayer. They reduced their own risk exposure of the GSEs risk exposure. They in essence diversify the sources of capital taking credit risk in the market. So they do help push us down a path of housing finance. And I think a lot of people have thought it was a wise path to go down, which is reducing the dominance of the GSEs dispersing credit risk, and market power and the like a little more broadly. So that’s sort of what CRT is there a bunch of different structures through which CRT is used, but by reducing the relief they get, they being the GSEs here to a level that probably puts this out of the money for the GSEs and all, but a few sort of points in the cycle.

Jim Parrot: It really does mean that the GSEs will revert back to a guarantee and hold sort of business the kinds of which they had before. So it really takes us back to a world in which the GSEs are less than intermediary in the market, which is kinda what they become in the last few years and more a just giant source of credit risk and market power and the way that they were heading into the last crisis, and it’s a little hard to get your head around because we’ve been trying now for years to disperse that concentration of risk and market power and to see a proposal that would reverse that it’s just a little hard to understand what’s going on.

Kirk Willison: So if CRT is devalued, does that mean that American taxpayers are at greater exposure for losses from Fannie and Freddie?

Jim Parrot: Yeah, for sure. In the analysis that we did in our paper we found that the GSEs might lose as much as 15% of their market share two thirds of which would go to banks, a third of which were going to FHFA. So you would think, I mean, normally when folks argue that GSEs have reduced market share part of the argument about reducing taxpayer exposure. But, because of the CRT treatment, and one other variable, which I mentioned, we actually found that the aggregate credit risk of the GSEs at the end of the day would go up. So you’re going to have a smaller footprint which has its own market problems that we can talk about it if you like yet a bigger risk exposure for the taxpayer.

Jim Parrot: What most folks have argued, well I certainly believe now is it makes sense to have a pretty broad government footprint cause you provide much more stability to the market,  continuous liquidity through times of stress or all kinds of reasons that the government footprint is helpful, and then you solve as much of the credit risk within that footprint, as you can, so that you get both stability cause the footprint and a more modest risk exposure for the taxpayer. For some reason, FHFA revision is going in exactly the opposite direction, which is to reduce the footprint and all the stability that comes with that yet increased the exposure to the taxpayer. So it’s pretty counterintuitive to say the least.

Kirk Willison: One of the things that we always associate with Fannie and Freddie are affordable home loans. What’s going to happen to the cost of mortgages under this proposal?

Jim Parrot: Yeah, so we found that the mortgage rates would go up anywhere from 15 to 35 basis points depending on the cycle, depending on where they are in the cycle and whether they get out of conservatorship, which is another variable.

Kirk Willison: You suggested this proposal will lead to smaller GSEs. What are the policy implications of that during say an economic downturn?

Jim Parrot: Potentially pretty dramatic. If you think about what we’ve seen since March and the economic fallout from the pandemic, if we’ve learned nothing else it’s that in times of stress liquidity functions in the government footprint and largely doesn’t function outside the government footprint. We had in the dark days of 2008-2009 lending outside the government footprint froze, I mean, all but stopped, but lending inside the government footprint, especially in the GSEs were put in conservatorship was maintained. It was a lifesaver in a sense same thing’s going on now. I mean in March the world outside of the government footprint all but shut down. And yet we’ve had this remarkably robust refinance market within the footprint and home sales to some degree, once people got to come out of their house and look at homes within the crisis in ways that make no sense given the underlying economics and given the disruption you’re seeing all across the economy, the degree to which the housing market’s held on is in large part because of the government footprint.

Jim Parrot: If you didn’t have it, if it were much smaller then the level of disruption would be undergoing right now would be a lot higher. So we’re setting ourselves up with a much smaller footprint. We’re setting ourselves up for a less stable system in downturns, less liquidity in downturns, and then there are all kinds of other more difficult to quantify implications. The cross subsidy for instance that the GSEs use to lower the cost for underserved communities it comes from a broad footprint. It comes from overcharging those that are lower credit risk, and if you shrink the government footprint and begin to lose those low credit risk bars to banks, you’ll lose that source of cross subsidy. And so, you know, MI folks have to pay more. So they’re the ripple effect of a smaller government footprint is it’s pretty considerable.

Kirk Willison: Well, Jim Parrot, I am so appreciative of your willingness to take your time, offer your expertise to us as we kick off this very first Arch Mortgage Insurance Policy Cast. I can’t imagine a better guest to start us off with. Thank you very much.

Jim Parrot: Really appreciate it. I’ve enjoyed it.

Kirk Wilson: Great.