As Prepared for Delivery
The second quarter was another very good quarter for Freddie Mac. Our net income was $2.5 billion and our comprehensive income – which is the number we most focus on while in conservatorship – was $2.4 billion. This included about $264 million after-tax from a good outcome in litigation involving certain non-agency mortgage-related securities.
Our quarter reflects a combination of strong business results and specific strategies to transform those operating results into strong and stable financial performance. This quarter, my comments will mostly focus on the latter topic – that is, how we are building a franchise that produces strong and stable financial results from the underlying operating performance of our guarantee businesses.
To start, let me say that there are three broad themes that are the pillars of building a long-term track record of such strong and stable financial performance:
Our growing track record of strong and stable financial performance is producing, in turn, a solid foundation upon which we can further prioritize our mission, our customers and innovation to create a better Freddie Mac and a better housing finance system.
So, first up is Good Earnings Stability.
Last quarter I spoke to you about the increasing stability of our earnings. That stability is no accident. It's the result of conscious decisions we've made in all three lines of business and at the corporate level.
We pursue this objective at two levels: First, tactical, where earnings can move significantly based upon fair-value accounting for market-sensitive aspects of our business. And second, strategic, since our earnings are very much exposed to the mortgage credit cycle.
On the tactical side, this speaks to actions that include:
While our earnings will always be sensitive to fluctuations in market conditions, especially given fair value accounting, we are now developing a track record of containing the net impact of these fluctuations to modest levels. Over the last six quarters, for example, the net impact has averaged just $125 million after tax.
On a strategic basis, we face the mortgage credit cycle full-on given our monoline charter and mission. But we have most assuredly learned from the Financial Crisis. We began years ago to emphasize credit risk transfer as a core part of our business model, first in Multifamily starting in 2009 with the advent of the modern K-Deal and then in Single-Family starting in 2013 with the first STACR bond issuance. And to put it bluntly, we have pioneered almost every aspect of GSE CRT.
To date, the objective of using CRT has mostly been to reduce the need for modeled capital required under the "Conservatorship Capital Framework" – I will call it simply "CCF capital required," – which is in turn tied to a severe adverse stress scenario. More about that later. But we are now also beginning to ramp up using CRT to reduce our exposure to credit losses in more routine markets. I will also discuss this topic in more detail a bit later.
Moreover, with CECL approaching – that's the new FASB Current Expected Credit Loss accounting to be implemented in 2020 – credit provisions will become more market sensitive each quarter. As a result, our efforts to reduce exposure to credit risk in routine markets via CRT will become particularly valuable to address this new source of potential GAAP earnings instability.
The second pillar of our financial results is delivering Strong Credit Quality
Our credit quality today is strong in both our mortgage businesses. This is a testament to a combination of factors all moving in the same positive direction – that is, specific business decisions we have made related to responsible lending, and to distributing credit risk to private investors, along with a very favorable housing market these last several years. I note that the favorable housing market itself reflects a strong economy – with a low unemployment rate as one measure. It also reflects a strong housing market, where limited production of new homes, among other factors, has led to steady gains in housing prices on a nationwide average basis, ever since the bottom was reached in 2011.
On the business side, we have been aggressive in:
Turning to our mortgage businesses, let me start with Multifamily. This business has a fabulous track record of low credit losses. This quarter, delinquency rates are down to just one basis point, along with currently zero real estate owned properties. This reflects not just a strong rental market – which is indeed quite strong – but also superior underwriting, along with a business model where that underwriting is NOT delegated. It's the apartment-house equivalent of kicking the tires with our own feet. And on top of this, Multifamily transfers about 90 percent of the credit risk on almost all new purchases – as measured by the reduction of CCF capital required. This reduces our earnings exposure to losses even more. And, almost all of these Multifamily risk transfers include credit loss protection starting from first-dollar loss.
The Single-Family business is naturally similarly focused on delivering strong credit quality.
First, in terms of legacy asset disposition, the real challenge has been to develop several different transaction structures to cost-effectively reduce our exposure to legacy single-family credit risk assets, which are indeed still significant. For example, we were the first GSE to sell non-performing loans, the first to execute a senior-subordinate structure to reduce credit risk on reperforming loans, and so on. In fact, such legacy-type assets are now down to only 33 percent of our mortgage-related investments portfolio, versus 58 percent just five years ago. We regard this as a great success.
Second, we have a strong, responsible, post-2008 credit box. It is, as I have already said, designed to be the consumer-equivalent of "investment grade," a term from the corporate bond markets. It's all very carefully monitored as to various measurements of potential loss statistics, for example.
For Single Family, to date CRT on the guarantee book has been almost exclusively driven by the economics of reducing CCF capital required. I remind everyone that the biggest cost component of the G-fee is, indeed, the cost of capital, so this prioritization was appropriate. We have more recently begun to test-drive CRT transactions that also transfer losses in more normal markets, rather than just severe stress ones. We are in consultation with the FHFA about expanding this program.
As a result, our serious delinquency rate is now down to 0.82 percent, the lowest since 2008, despite the hurricane losses of 2017. On the post-legacy book, it is just 0.25 percent, reflecting our investment grade-centric credit box combined with strong house price appreciation since 2011. We also then use CRT to additionally reduce our credit risk on new flow business – again as measured by CCF capital required – by about 60 percent. Further CRT innovations in the pipeline should increase that percentage.
I do want to state, as a major mortgage market participant, that credit risks are low due to, among other things, the underlying failure of the marketplace to produce enough new living units versus historic norms and current demand – whether they are stand-alone or multi-unit dwellings, whether they are owned or rented. This shortage creates upward pressure for both house prices and rents – both of which are good for credit quality in our business. But for the country and the economy overall, it is not so good after a point. In addition, the shortage is felt most acutely at the lower-price points, which speaks to affordability.
So, I worry that we are at a rate of house price appreciation that is above a good equilibrium and sustainable level due to this shortage of production. Freddie Mac, as a guarantor, is doing what it can to help address this problem – for example, by purchasing certain types of multifamily loans tied to rehabilitation and refurbishment – but we can only work at the margins by the nature of our Charter. Still, a broader response is needed, and I look forward to policymakers successfully addressing the issue, hopefully, in the not-too-distant future.
The third pillar that supports our financial results is the new business model we have built based on credit-risk transfer, which we have pioneered. If you look back historically, you will see it is in many ways an extrapolation of prior GSE business model evolution where the pass-through security was developed in 1971 to move interest rate risk and liquidity risk away from GSE balance sheets and to the capital markets. CRT now extends this concept to credit risk. The result is a solution that has transformed our business – and is changing the way a significant portion of the U.S. housing market is funded.
Our leadership in this space – shifting risk away from taxpayers in an economically efficient way – is a great source of pride for the company. These transactions are also providing new investment opportunities for investors and, we believe, more broadly strengthening American housing finance. And importantly, CRT investors are providing frequent market feedback on our credit quality and our credit risk management – in both the Single-Family and Multifamily businesses— so we have many sets of eyes providing risk feedback to us today.
By transferring risk away from our company in a responsible way that does not reduce liquidity or adversely affect the availability of mortgage credit, we are in turn increasing our own corporate stability through the credit cycle – in good times and also more challenging times. That stability is a key objective of ours, as per Freddie Mac's charter.
As some of you may have noticed, we surpassed a major credit-risk transfer milestone this quarter. Freddie Mac Single-Family has now transferred a portion of the risk on mortgages with an unpaid principal balance of over $1 trillion. Given that the market didn't truly exist as little as five years ago, that's quite an accomplishment – one that, we believe, has made Freddie Mac and the entire system far more safe, sound and stable.
The primary driver of much of our CRT activity, as I stated earlier, has been to reduce the CCF capital required to support our credit risk. Now that CCF has been established, it serves as a tool for communicating the quantification of that risk reduction – to our regulators, the markets and the policy community. Previously we could only rely on our internal economic models as a reference point. In addition, for good reason, we have found that the capital raised via CRT is inexpensive in comparison to our estimate of what the marketplace would demand if we held the credit risk, as was traditional pre-conservatorship. The result is to drive the underlying returns on our businesses higher on the reduced credit risk we retain. Today, in conservatorship, this means taxpayers are earning a better return on their support of the company. In a potential future state, it means capital would be easier and cheaper to raise.
As you may have seen, for the first time since conservatorship, our quarterly financials discuss return on CCF capital required, which we treat as a proxy for return on equity – ROE. Let me provide some color.
In 2017, the FHFA issued guidance to use the Conservatorship Capital Framework to evaluate and manage our financial risks and also to make economic business decisions while in conservatorship. This capital system is generally, as per the FHFA, broadly consistent with the Dodd-Frank Act Stress Test in its approach and stress assumptions. We now use this capital system to measure internal transactions and lines of businesses from a risk-return perspective. It provides us with a metric to determine whether a given transaction or some business line makes economic sense to do.
Since our support comes from the American taxpayer via the U.S. Treasury, we owe it to those taxpayers to make sure that we engage in transactions and run our businesses on a smart economic basis. So, for example, when we do a single-family CRT transaction, we compare the G-Fee paid away to investors with the economic value of the amount of risk reduction, which requires a capital system like the CCF to measure. That way we ensure we don't overpay for risk reduction.
We are always careful stewards of the taxpayer's support.
For the second quarter, adjusting for the significant item, we estimated that our return on this CCF capital required was 16.4 percent, representing an increase from 12.9 percent in the second quarter of 2017. We also note that these numbers are not fully indicative of what our ROE would be in some possible or likely future post-conservatorship state; our post-conservatorship ROE would likely be lower, we estimate, for a variety of reasons, possibly substantially.
At the beginning when I mentioned the objective of "strong and stable earnings," I did not elaborate on the definition of "strong." For a financial institution like Freddie Mac, the most common measure of earnings strength is, in fact, return on equity. With the CCF in place, enhanced by its inclusion in the FHFA's Proposed Rule on Enterprise Capital recently published in the Federal Register, we now feel comfortable disclosing this CCF-based proxy for ROE for the first time and, as you can see, it indicates that our earnings have gotten to the point of, indeed, being strong. CRT, as a core business model, has been key in delivering those results.
I would note, again, that the CCF-based proxy for ROEs is likely to be higher than an actual ROE in a possible future state because, as one example, most reform proposals call for us to pay an explicit fee of some amount for federal support to our balance sheet, where now it has no overt accounting expense.
Finally, let me conclude with this important point. The strength and stability of these financial results –as supported by these three interrelated pillars of (1) tactical and strategic earnings stability, (2) strong credit quality and (3) our new CRT-based business model – have given us the foundation necessary to grow our businesses and to focus on our mission and on innovation.
As a result, our origination volumes remain strong: Single-Family new originations this quarter were $84 billion, with purchase volume up 29 percent from the prior year, while refinancing volume dropped seven percent given the higher rate environment. Multifamily new originations were $16 billion this quarter, up 13 percent from the prior year. Our guarantee book continues to grow, moving up six percent year-over-year, to reach nearly $2.1 trillion – demonstrating how we are fully participating in mortgage market growth.
We also provided $103 billion of liquidity into the mortgage markets in the second quarter. That funded more than 362,000 single-family homes and 191,000 multifamily rental units. In Single-Family, our support for first-time homebuyers is at the highest level in the last ten years, at 46 percent of new purchase loans. Meanwhile, for Multifamily, 87 percent of the eligible rental units financed were affordable to families earning at or below area median incomes.
Without the underlying strength and stability of our financial results, it's difficult to see how we could be so strongly focused to make this kind of progress in accomplishing our mission.
Thank you for joining me to discuss Freddie Mac's second quarter financials. I look forward to your questions.