Description

Long Genworth (GNW) subordinated floaters at 66% of par (37247DAG1)

Genworth Financial is the former insurance business of GE Capital and has been one of the more controversial financial stocks in the US over the last 20 years. Alternatively viewed as a cheap long-dated call option on rising interest rates or as a dumping ground for GE’s toxic long-term care (LTC) insurance policies, the company has arguably attracted outsized attention by investors over the last 15 years compared to its size. There have been 8 different VIC writeups on GNW over the years, which provide great context and background information. After several near-death experiences and a failed merger, Genworth has managed to limp its way forward albeit only into a state of market apathy. After some exciting twists and turns, the prices of the company’s securities have finished the journey where they started and trade in the same rough context as they did a decade ago.

The main point of this writeup is to argue that the turnaround seems to have basically worked: Genworth’s equity now has obvious value given its ownership in a publicly traded mortgage insurer (Enact or ACT), and the cashflow to the holding company has dramatically improved since the bad old days of 2016-2021.  We further argue that subordinated floaters offer a very good risk/reward proposition since one of the primary uses of that cash flow will be to retire holding company debt, including the bonds in question. Since the market value of the ACT stake by itself is higher than the market value of all of Genworth combined, the bull case for GNW equity is that the company will spin out those shares and collapse the valuation discrepancy. We believe that the spin of ACT shares is a likely outcome in the medium term, but we also argue that substantial debt retirement will need to occur before Genworth’s “RemainCo” is a viable standalone entity. Consequently, we expect the bonds to receive a lot of the cash flow accruing to the holding company, resulting in the company repurchasing them well before their stated maturity. The sub floaters specifically are priced on a yield-to-2036 basis and can generate equity-like returns if they are substantially retired over a 4 year period.

While Genworth’s corporate structure historically was extremely complicated, today it is very simple:

• Ownership of 131mm shares of publicly traded mortgage insurer ACT (worth $3300mm as of writing)
• Ownership of regulated life insurance and LTC subsidiaries (assumed to be worth zero)
• Ownership of a startup providing fee-based PPO-style provision of care in the LTC market (Care Scout)
• Litigation claim against Banco Santander, via Axa insurance ($830mm face value)
• NOL related tax receivables
• Net debt of 643mm

As noted earlier, the bull case for GNW equity is that the value of the ACT stake currently trades for more than the value of GNW as a whole. For its part, ACT’s share price is depressed due to low trading liquidity resulting from GNW’s majority ownership. The obvious solution is for GNW to spin out its ACT stake, collapsing the discount and allowing ACT shares to rerate higher on more normal levels of liquidity. The primary issue is that the remaining assets at GNW would be a startup (Care Scout) and regulated life insurance subsidiaries with trapped capital.  Care Scout, by itself, will not be in a position to service the current level of debt at the holding company and dividends from ACT are the main source of cashflow today. Therefore, a precondition to a successful spin of ACT is for GNW to retire the majority of its outstanding debt beforehand. Among other considerations, there are obvious fraudulent conveyance claims if Genworth were to spin out the ACT stake and then default, not to mention default being a completely unnecessary hypothetical at this point.

Luckily between NOL utilization and ACT dividends the Genworth holding company is awash in cash flow. The cash from NOLs is relatively straightforward: the life insurance subsidiaries generate taxable income, which regulators allow to be up-streamed to the holding company to pay taxes on the group’s behalf. The holding company has net operating loss carryforwards which shield the actual tax payment, allowing the holding company to retain the cash to the tune of 150mm remaining in 2023 (with 50mm already realized in Q1). For its part, ACT generates around $550mm of net income each year and starts with a $2100mm PMIERs regulatory capital surplus. ACT’s insurance in force has been growing in the mid to high single digits due to persistency, and the existing regulatory surplus is more than sufficient to support growth in risk in force (and earnings power). When Genworth’s holding company was in more dire financial circumstances a few years ago, regulators imposed capital restrictions on ACT that limited capital return out of fear that Genworth might raid the subsidiary for cash. Given Genworth’s recovery, these capital restrictions were removed on March 1st 2023. We estimate that total payout ratios can rise from 50% to as much as 75% (~300mm+ per year to GNW) and still grant ACT management excess capital for any growth initiatives.  Furthermore, in 2024 ACT begins to release contingency reserves that it started building in 2014, allowing the company more flexibility to distribute earnings at its discretion. For context, mortgage insurance peer Radian has returned ~75% of its earnings to shareholders the last few years.

Regarding the stability of ACT’s dividends, we would argue that Enact’s mortgage insurance earnings are likely to be way more resilient than seemingly implied by market multiples – not only for Enact itself but also for the mortgage insurance industry in general.  The first main point is that mortgage lending standards since the GFC have improved dramatically: the advent of Qualified Mortgage lending standards now explicitly exclude loans with the characteristics that generated the majority of losses in the last cycle (income and assets are verified by definition). The second point is that the average FICO of ACT’s risk in force today is materially higher than it was in prior cycles. The third point is that changes in GSE loan servicing policy now prioritize forbearance measures when borrowers get into trouble. Mortgage insurance only pays out when lenders a file a claim on foreclosure, so the widespread use of forbearance increases cure rates and diminishes ultimate losses (e.g. delinquencies are less likely to turn into paid losses because the borrower has increased time to catch up on payments). Finally, today Enact systematically utilizes reinsurance coverage to protect against catastrophic credit loss, mitigating credit volatility. While reinsurance lowers the company’s return on equity it increases the stability of net income by damping tail loss outcomes. Granted, we think it’s unlikely that the market will give credit for these changes in the mortgage insurance industry via higher multiples until the after the next recession has passed without serious problems, so ACT’s valuation may well remain suppressed for a while. Yet from the perspective of the bonds the primary concern is not Enact’s share price but rather its realized losses and dividend paying capacity, which should remain very substantial. In the event that our dividend forecast proves too optimistic, we believe that Genworth would prioritize cash received for debt repayment rather than share repurchases.

We think that the Genworth holding company will receive up to $1.2bn in cash cumulative through 2026 for use in both debt and equity repurchases. This does not include any cash generated from a possible or even likely legal settlement with Axa/Santander over that time frame. Given Genworth’s multiple near-death experiences related to liquidity over the last decade and the fact that Care Scout will likely not be able to support much leverage for medium term, we believe a very likely outcome is that Genworth uses that cash to retire both series of bonds to near zero outstanding.

There are two bond issues outstanding:

Senior unsecured 6.5% due 2034: There are 276mm outstanding after the company repurchased 13mm principal in 2022 and 11mm in Q1’23. These bonds benefit from a Replacement Capital Covenant implemented in 2006 upon the issuance of the junior subordinated notes. The RCC specifies that the company cannot repurchase or retire the junior subordinated notes unless they are replaced (refinanced) by similarly junior securities or equity as long as eligible debt is outstanding. The definition of eligible debt requires outstanding principal of $100mm or greater. Hence, as and when Genworth retires this issue of bonds to less than $100mm outstanding, the company can start repurchasing the junior floaters.

Junior subordinated 3mL +200bps due 2036: There are 600mm outstanding with a scheduled maturity date of 2036 and final maturity date of 2066. On the scheduled maturity date in 2036 Genworth is required to repay these bonds using the proceeds from a new issuance of “replacement capital securities” (essentially junior subordinated debt, mandatory converts, or equity) subject to commercially reasonable efforts. This replacement capital obligation is essentially a bull market legal clause; when Genworth was facing liquidity concerns 5 years ago this clause was irrelevant for all intents and purposes but now that the equity case has significant merit the language motives Genworth to retire the bonds early rather than face the prospect of issuing (high coupon) preferred stock or common equity.

The path forward: GNW holding company generates significant cash in 2023-2026 from subsidiary tax payments / NOLs, ordinary and special dividends from ACT, and possibly from a legal settlement. Genworth will likely continue repurchasing the 2034 bonds in the open market until there are less than $100mm outstanding. Once these bonds no longer qualify as eligible debt under the RCC, GNW should begin to repurchase the subordinated bonds in the open market. We would argue that Genworth will likely retire both issues to near zero outstanding rather than burden the Care Scout growth venture with significant interest costs. The significant deleveraging at the holding company likely results in ratings upgrades and spread compression for the bonds. 

Addendum:

What if GNW decides not to spin off ACT? Most of the argument for the subordinated bonds hinges on the idea that Genworth needs to retire them before spinning off ACT as CareScout cannot support the debt load. It’s certainly possible that Genworth management decides to retain ACT and its cash flow, which reduces the need to retire the debt. In this case the bonds should simply trade on a YTM basis, in this case being the stated maturity of 2036 as it’s hard to see the company wanting to trigger the replacement capital obligation. However, retaining ACT would cause cash to pile up on the balance sheet, which would be a positive for spreads. We would point out management’s commentary regarding this topic in the May 2022 earnings call when it noted the following:

“Longer-term, after Genworth addresses its remaining debt maturities, brings the legacy LTC portfolio closer to economic breakeven on a go-forward basis, and makes headway on our LTC growth strategy [CareScout], Genworth would be in a position to consider other options for Enact including a spin-off of Enact shares to Genworth shareholders”.

We would expect GNW to be in a position to have met all of the criteria for the spin by 2026. The spin is not only what equity holders in Genworth have been demanding but also the most obvious means of generating share price performance (TSR is a component of GNW management bonus calculations)

What about the LTC business? In full disclosure I’m old enough to remember when the company brought out an army of actuaries on stage at an investor day in 2013 to argue the adequacy of the reserving in these blocks, and then catastrophically imploded 12 months later because every single assumption presented that day was wrong. A decade later Genworth now estimates that it will have needed approximately $30.3bn of cumulative net present value premium increases for those blocks alone (2x the 2013 company-wide GAAP equity base), so perhaps “wrong” is an understatement. In 2015-2018 it was far from clear that state insurance regulators would actually grant premium increases of a magnitude sufficient to close the gap to breakeven, but so far they have granted $23.8bn worth of premium increases and continue to approve additional hikes every quarter.

• The company thinks it can close the $6.5bn NPV gap to economic breakeven within ~3-4 years. The primary driver to close that gap are additional rate actions that increase premiums on the order of $250mm per year, which translates to around $1.5bn NPV benefit per year. Genworth is also applying savings generated from their new CareScout preferred provider network to their existing LTC blocks to reduce claim severity on the order of $1.0-1.5bn NPV. Finally, Genworth is also seeing a reduction in claims tail risk (lifetime benefits, 5% compound inflation riders, spousal benefits) from legal settlements on their LTC policies which cover 80% of their in force lives. Under those settlements policyholders choose reduced benefit options (RBO) or non-forfeiture options (NFO) in exchange for lower or no premium increases, resulting in possible NPV benefits of $1.0-1.5bn from here.
• Even if the three mitigation strategies noted above prove to be insufficient (and there is no reason to suspect that), peak claim years on the LTC blocks are still 10 years away. This timeline gives the company plenty of opportunity to work things out although we think it’s probable that they’ll get there within a few years.
• Finally, we would note that Genworth cannot be forced to inject capital from the holding company into the regulated life insurance subsidiary (GLIC) and it’s nearly impossible that they would voluntarily choose to do so. However, it’s also unlikely that they will be able to extract any of the $3bn in statutory capital in those subsidiaries either, as that capital supports policy holders in the event that LTC forecasting errors continue.
• What about office commercial real estate? Despite the widespread assumption that GLIC’s value is zero, the market continues to trade GNW equity on worries about the performance of GLIC’s investment portfolio. Genworth has $1.8bn of loans on office real estate in GLIC. Unlike commercial real estate loans typically found in CMBS, Genworth’s exposures are very granular ($5mm on average) with amortizing principal payments as opposed to interest-only payments. 1/3 of the office loan portfolio is fully amortizing within the term of the loan and 2/3 are partially amortizing, leaving a smaller balloon payment at the loan’s end. Given good debt service coverage ratios including principal amortization, Genworth is likely to simply extend any loans that cannot refinance at maturity and receive paydowns via the monthly payment over time rather foreclose and sell assets. 

Catalyst

Dividends from ACT, open market repurchases, Santander/Axa litigation outcome