tv Urban Institute Discussion on Government- Sponsored Enterprises - Panel 2 CSPAN June 11, 2019 7:39am-8:42am EDT
>> senator cruise talked about the senior status of mister biden and mister sanders and said watch out, jimmy carter is still around. >> thank you, everyone. very good. thanks, everybody. >> the house debate a resolution today given the judiciary committee approval to sue attorney general william barr and for white house counsel don began in federal court to compel them to provide congress with information related to the mueller report investigation. the measure also lets other democratically controlled house committees sue trump administration officials for testimony and documents as long as they get approval from a bipartisan house leadership committee. watch live coverage of the house debate starting at noon eastern on c-span. watch anytime on c-span2 and listen on the free c-span radio apps.
now a look at the impact of freddie mac and fannie mae on the 2008 housing crisis and the current role in the housing and mortgage markets. the are been institute in washington hosted this one our discussion. [inaudible conversations] >> the second panel, we have panelists talking about this version of risk. i know the end of the last panel, the first panel they got a little bit of a credit risk so they tried to take our territory, this one is better so everyone should understand that. i will introduce the panelists first. going from my left to the far left.
the global mortgage group in orange capital, and the vice president have housing finance policy and david finkelstein. we start with laurie, through the data on history and development in credit risk transfer. >> it is a job of mentioning, started in 2012 and the idea is to reduce the overall risk of fannie and freddie, reduce the risk to the taxpayer and put
private capital first and basically grew enormously so there was $90 billion of collateral cover in 2013. $420 billion and the last couple years, 2017-2018, it has been $690 billion in 2017. and 18, altogether, $2.7 trillion of the exposure has been covered at least partially through credit risk transfer. and a lot of expertise and
analyzing credit risk for securities that have been at risk. they have the capability at the time they were watching ends disappeared, largely disappeared and legacy trading and volumes breakdown, the expertise was there so hani and freddie provided a tremendous amount of low-level data, allowing investors to understand the risk of the securities they were buying and the transparency and that provision of data with the success of the market. and the transfer risk has combined risk of $91 billion, of the unpaid ballot paid.
this doesn't tell the story. there has been a huge development in terms of this market and the diversity of structures. it started with the, and 30 year mortgages. covering the risk of loans, already in portfolio, with risk transfer to the capital market. and the range of structures to include mortgages with higher mortgages, 82, 97 to include
more variations in terms of attachment and detachment point with the change in structure over time. with capital markets transfer as well. capital market transfer, reassurance transactions, on loans already in portfolio to insurers and reinsurers, with access deals and a small amount of back end lender risk sharing. 2013-14, 90% of the transaction, capital markets transaction, and the ballots were institution based transactions.
and the amount of transactions as defined for the last few years, around 64%-65%, the back end institution that has grown to the mid-20s and you will notice freddie and fannie have introduced blunt institution based transactions, laying off the risk at the point of origination, this includes, this includes lender recourse deals, also includes deals with insurers and reinsurers. that is up to 12.5% of total origination. if you look at 2018 freddie and fannie credit risk, it is 65%,
other back end capital market transactions, 23% institution based back end transactions, 12.5% front end institution based transactions, very interesting to look at the diversity of structures is very important. of each of these structures get strengths and weaknesses. it worked late in good times but spread may be volatile providing less protection for the taxpayer but they do provide transparency to the market for the reinsurance lender recourse. front end risksharing provides greater risk of pricing and excess of what fannie and freddie would otherwise charge that can be managed as well as a low level of credit risk
which can also be managed. lender recourse is available only to large institutions. every one of these structures, by trying out all these structures you can figure out, how to have more strategic deployment over time. the emergence of credit risk transfer has really been one of the highlights of the conservatorship, as mark mentioned earlier with risk to the taxpayer, considerably considerably considerably. >> we will go back with laurie and keep going, and -- we start
with data. the area of responsibility encompasses that and arch in particular is a key source of innovation in the industry so i will ask you to talk about that. >> that background is important, we are both a key provider of credit risk transfer in the form of primary am i, for transfer structures for crt transactions to transfer credit risk on our own portfolio to third-party so we have a good perspective on all parts of the transition. both the buyer and a seller.
and historically cit existed in the form of private insurance. that was crt. it is an alternative but that was the first form of private mi and the first form of crt was private mi and we know how that ended up. coming out of the crisis the private mi had a business model that was buy-and-hold of credit risk transfer. they had the same cycles to our business, neither one has appropriate risk management techniques to diversify risk and that is how we got to the third crt structure looking for a method to diversify risk, transfer credit risk so gsc were not the only ones holding it. we have a robust credit risk transfer environment that spends capital market and insurance and reinsurance.
mostly on the back end basis but a little bit on the front end basis. i will come back to the front end a little later and why that is so important. another issue, what is the permanence of crt. and that comes down to private capital would be available if private capital feels it is getting an appropriate return for the risk it takes. so long as the question is answered private capital will exist and private capital will continue to support the crt function and potentially break down when that return is not there and private market feels they are not getting what they need to. it is important that we recognize that and there is a lot of permanence there and
equity-based capital and structure based capital, what it is down to is the return for private capital. whatever form the transaction takes to me is a lot less important than the return they are getting. it is critical to getting us to where we are today. you can't accomplish innovation without experimentation. can't have innovation without trying new things and to try new things you have to have experiment and pilot programs. some ability for the gscs or whoever is looking at participating in the market has risk or someone looking to purchase credit risk protection to try new structures and try new things and it becomes dangerous if we try to restrict that because we are stifling innovation and not allowing the
market to develop and get to where it needs to be. lori did a great job pointing out how the market evolved from being almost entirely institution back end capital to institutional capital as well as insurance and reinsurance capital and from back to front end. one of the real critical elements of the front end is feedback. when you think of the gscs, fannie and freddie, using my previous contention will be there for the transactions where they think they are paid for the risk. when you see private capital pulling away, a strong message allowing them to react in a certain way. at the front end that is more timely feedback and you are not left with large popular
exposure. front end structure is incredibly important and for that to succeed you need experimentation and that means you need pilot programs. that will be successful and some will not be successful but ultimately we need that as part of the business development. as we cycled back we have a lighter point. >> a keyword that we will come back to, good times and bad times, different structures. i will go to david finkelstein, what this means for markets, between you and mike and the others, not for market investors but homeowners and families.
>> what david discussed, the evolution of the crt market. whether this would really book and 6 years later it has been a significant success. it capitalizes on the core competency, private capital can evaluate the credit risk, government can provide catastrophic risk. it is distributing the credit risk. it has been a good experiment. it will continue and expand. and ways to make it more durable in terms of protecting taxpayers but these come at a cost.
there were three potential ways to enhance the program. what this was talked about in policy circles, how much risk potential. there has been some arguments that it should be higher and looking at every objective model through the crisis you are well protected in the context of 4%. to the extent there is desire to go higher up the capital staff and relying on private capital to take that catastrophic risk, private capital with higher cost of capital than the government, you will pay a premium for that and if you were to go to 10%
the incremental costs of buying that insurance would be 8 basis points or absorption in the mid 50s. you don't get much for it but the costs are increasingly burdensome. another mechanism is how to make execution more predictable for the gscs. it is the case that crt spreads can be quite volatile and once there is a victim of success, the markets develop, they become somewhat of a benchmark for residential credit and as a consequence of that, the sector is often influenced outside macro factors that influence other markets, they are a benchmark and spreads can be
erratic and that puts signal risk on execution given these are back end and execution is based on production from four months ago so that is a problem. one thing the gscs might do to mitigate that risk is create structure which is common in other markets namely commercial and corporate markets where you replace runoff in a particular crt structure with new origination that exhibits the same credit characteristics so say for example at the outset of the crt issue, for the next three years, loans are paid off, gscs replace those loans into the structure without having to sell new crt at the spreads at that moment. that will mitigate a lot of the
pipeline risk but it is not a cheap endeavor to do the private market. and will charge you probably roughly 100 basis points across the first 0%-4% of that structure and that will equate to 10 to 12 basis points of absorption and that is probably pretty significant. but it is a considerable risk mitigate or when it comes to execution. given the fact that when you look at execution of crt in the current market there is roughly 20-24 basis points of compensation left over after expenses, after payroll tax and after crt execution of the current market so you can make an argument that 10 of 12 basis points might be a viable investment on a portion of the
crt execution, not a significant portion but to offset the risk so it is certainly something to consider. another potential path is to sell more first loss. initially six years ago the market wanted the gscs to have skin in the game and so it was recommended not to sell first loss. .. but given the fact markets or spreads are quite variable, there are opportunities to sell
first loss risk when it's optimal to do so like in the current environment where spreads are relatively tight and you can lay off some of that execution or so met first loss at economical levels. i would say that something the gses should think about episodically in terms of enhancing what they're disturbing to private markets. up the stack, wouldn't recommend, too expensive for what you get. a revolver type structure were you able to replace runoff. worthwhile investment for portion of the loans and something that should be considered and then episodically selling more first loss. but again these do come at a cost, and when profits become less available in times of markets are not as good, it's going to push the gses to limits. that was just on gas and stucco.
we would think about the private types of transactions that are current on the market we are supportive of those being more common for the exact reason that spread variability can post execution risk. i would think about it, however, is that normal functioning markets, capital markets execution given liquidity and the transparency is always going to be the lowest cost execution. markets are not always normal functioning obviously and so when you have the ability to engage in transactions that are less focused on capital markets, you tend to see spreads or costs that are little bit more stable. we see this everyday in our business. with three credit businesses, each with direct lending platforms including residential home loans, and the sectors we see day in and day out, they are a step removed from capital
markets embassy much more stable pricing that would definitely supportive of the gses further pursuing these tracks. now, you are a couple of caveats. number one is obviously they are more expensive to execute. the baseline price is going to be higher. we see it in crt for example, where the backend tranches are roughly five five basis pointse in terms of gt that acosta lip that risk when we normalize the attachment. it you want front end whereby you can lock in some flow and there were deals done as recently as last year where icing was locked for roughly six months and going to cost you an instrument of three basis point or thereabouts. all in about eight basis points. your terminal at risk in terms of managing a pipeline interface
riddled execution risk over the not-too-distant future, six months thereabouts. it's something we would recommend. the lender specific transactions are good. we've actually explore them ourselves. we negotiated where we would like pricing for a year as well as falling with one specific lender to one of the gses. however, again, with the starting offer on that transaction will be 50 basis points above cas and stacr execution. we did not agree to those terms. if you are let's see what a basis point to cost you about three or four basis points in additional g-fee or absorption. something we would consider another point with the lender specific transactions and you could potentially face adverse selection in the event that became much more robust.
there are, in spite of the fact credit may look very similar across the gse bucket, probably, different aeruginosa have different credit card receipt. originators with much faster prepaid came to a better credit because there's less opportunity for life events and defaults. and so you could see specific lenders with better credit channeling their crt through these transactions, and then you end up with the worst credit lenders going through the cas and stacr market which ultimately would be born the cast would be born by the gses because we're going to charge more for that. so i'll leave it at that great. thank you. again i want everyone, i want to turn to mike and laurie next, but just for and what else to think about, i was interested in the idea of the market not
insisting on the skin in the game from wanting the first loss credit risk which is wrote an interesting development. so later on any thoughts on that. why don't i go first to mike and then -- take it away. i know your thoughts on all of this, and the note you will talk about what it means for families and homeowners in the market. >> so today we been talking about gse reform and there are several different layers, if you will. a lot of what we've talked about is the plumbing, how many gses, how their organized, relate to each other. and above that is what the regular with you, conservative like powers, how will that be preserved after conservatorship. i think everybody generous supports a very robust oversight from the regulator. that's one of the reforms in