Risk/Reward of BDCs in Income Portfolios – Systematic Income Investing

Income investors have a number of different types of investment vehicles at their disposal. In this article we take a look at Business Development Companies. In our view, it is less useful to look at a particular investment structure on a standalone basis and, instead, more useful to highlight the key differences between competing options. The most direct alternative for BDCs are credit CFEs so in this article we compare and contrast BDCs to credit CEFs along a number of different dimensions. The main takeaway is that BDCs look attractive relative to credit CEFs across a number of key factors and are worth a look.

We also highlight the Fidus Investment Corp (FDUS) with the aid of our recently launched BDC Investor Tool which can help investors to cut through a lot of the noise and focus in on the most attractive options in the sector.

Income Visibility

A key issue for many income investors is the visibility of income from their investments. Here, BDCs come out ahead of CEFs because they tend to provide greater transparency about their underlying income levels. First, BDCs report earnings on a quarterly basis whereas CEFs issue shareholder reports only semi-annually. It’s true that a number of fund companies release earnings on a monthly basis but they are in the minority.

The second income transparency advantage of BDCs has to do with the nature of income produced by CEFs and BDCs. BDCs tend to have primarily floating-rate assets whereas credit CEFs run the gamut from primarily fixed-income (i.e. preferreds, bonds) assets to primarily floating-rate assets (ABS, loans, CLO mezz etc). We have made the point repeatedly that fixed-income CEF distribution rates tend to run well above their portfolio yields because of the simple fact that the currently low level of interest rates has pushed most fixed-income securities to trade well above par causing their current yields to exceed their yields-to-call and yields-to-maturity. BDCs don’t have this problem because floating-rate assets don’t move much above par and so do not feature a negative pull-to-par dynamic. In other words, BDCs offer a what-you-see-is-what-you-get kind of income, whereas most credit CEFs make it seem like their yields are higher than they are in reality.

Yield Level

The second key point for income investors is, simply, the level of yield on offer in credit CEFs and BDCs. The level of yield between the two types of vehicles is determined by the yield of the underlying assets, level of leverage and fees. As far as the level of underlying yields, BDCs tend to enter into secured loans with yields typically in the range of 7-10%. Credit CEFs, on the other hand, have access to assets with yields in a lower range of 4-6% based on current valuations of bank loans and high-yield bonds.

BDCs also benefit from higher leverage levels. It can be tricky to compare CEF to BDC leverage because the definitions used across these two types of RICs are different. CEFs use the definition of debt / total assets while BDCs use debt / equity. So, a portfolio of $200 total assets and $100 of debt has CEF leverage of 50% and BDC leverage of 1x. Credit CEFs tend to run at lower levels of leverage of around 30-35% (BDC-equivalent of 0.5x) versus BDCs of around 1x (CEF leverage equivalent of 50%). A difference between 35% and 50% may not sound like much but it is quite substantial. Leverage of 50% means investors have exposure to 50% more of assets on a net basis than leverage of 35% i.e. assets of $200 for $100 of net assets versus just $154. Higher underlying yields and higher leverage allows BDCs to generate more income on a net asset basis than credit CEFs, all else equal.

On the fee part of the equation, CEFs come out ahead with average fees of sub-1% whereas BDCs tend have base management fees of 1.5-1.75% with additional income and capital gains incentive fees above certain hurdle levels. Relative to underlying yield and leverage levels this has a more marginal impact.

If we estimate NAV yields based on these figures we get to a number of about 5.2% for a credit CEF and 8.5% for a BDC. It’s not a coincidence that 5-year total NAV returns for loan CEFs and BDCs have roughly reflected these numbers at around 5% and 9% CAGR on a median basis respectively though there is significant dispersion around these numbers.

An important additional point here is that the higher underlying yield profile and higher leverage of BDCs versus credit CEFs should also lead to a higher default and loss rate for BDCs on their holdings. That certainly does happen on an individual basis as there are clear underperformers in the BDC space but on an aggregate basis it appears to be a wash i.e. BDCs in our 25-company sample have outperformed loan CEFs roughly by their net yield advantage.

A key consideration for investors is diversification and here CEFs come out ahead. It is much easier for credit CEFs to build a well-diversified portfolio. This is because their holdings trade in the market and can be sourced with relative ease, if not immediately, then over days and weeks for the vast majority of typical credit securities held by CEFs. Many are not very liquid, certainly, but there are enough issuers with sufficient liquidity to allow dozens of credit CEFs to build portfolios with a number of positions in the triple digits. It is harder for a BDC to build a similarly sized portfolio as nearly each transaction requires a search process, negotations with the borrower, bespoke documentation and more. This leaves some BDCs less diversified with “chunkier” portfolios than their CEF counterparts.

Deleveraging Risk

An additional consideration for investors between the two types of investment vehicles is the risk of deleveraging. A key driver of CEF performance, particularly over 2020, was whether or not a given CEF was forced to deleverage or not – a topic we have discussed a number of times in the past. With respect to deleveraging risk, there are four key issues to consider – asset coverage rules, any additional requirements of financing instruments, any additional leverage mandates, and how the NAV is calculated.

Generally speaking, the leverage provisions of the 1940 Investment Company Act do not apply to CEFs because CEFs don’t finance themselves with senior securities. Rather, they use derivatives (such as CDS) or bilateral bank-facing instruments such as repo or credit facilities. BDCs, on the other hand, tend to finance themselves with unsecured debt and, to a smaller extent, with secured credit facilities, and occasionally, with securitized vehicles such as CLOs. This may seem like a win for CEFs, but it’s not because bank-facing instruments are daily marked-to-market and give the lending bank the ability to liquidate the collateral it holds if, in the view of the lender, its value falls below the pre-negotiated value. BDCs, on the other hand, are simply prevented from carrying out certain actions such as paying common dividends, issuing new debt, etc. if they fall below the regulatory asset coverage guidelines. In other words, unlike CEFs, BDCs don’t tend to be in immediate danger of fire sales on asset coverage breaches.

Separately, the SEC has shown its willingness to relax asset coverage rules during a severe market dislocation such as the one we saw in 2020. Specifically, the SEC allowed BDCs to adjust asset coverage ratios lower last year to enable them to continue to put capital to work.

Perhaps more importantly, is the fact that BDCs hold non-traded assets and are not as anchored to market valuations. For instance, one would think that because BDCs run at higher leverage than loan CEFs and they tend to hold higher-yielding (and, hence, arguably, lower-quality assets) their NAVs should have taken a bigger hit than the NAVs of loan CEFs over Q1 of 2020. In reality, the average BDC total NAV return over Q1 of 2020 was a drop of 9% while the average total NAV return of loan CEFs over the same period was a drop of 30%.

A drop of 9% means that, on an unleveraged basis, the BDC portfolio performed roughly in line with investment-grade corporate bonds over Q1 of 2020. This is obviously ludicrous and there is no reason we should believe this valuation is “real”. That said, the ability of BDCs to dampen moves in markets with their valuation assumptions given them a kind of superpower. Investors familiar with private equity companies are well aware that private equity companies have a similar sort of “exorbitant privilege” which allows them to benefit from leverage without being as exposed to increased deleveraging risk from market moves as investment vehicles that hold traded assets which they have to mark-to-market daily.

The combination of leverage and market pricing is a dangerous one as investors in agency-focused mortgage REITs are well aware. Over Q1 – the average agency-focused mortgage REIT dropped about a quarter of its book value on average despite agencies rallying over the same period. The dislocation in markets (specifically, the widening of the basis between agencies and hedges) combined with a reliance on daily mark-to-market financing instruments like repo caused a wave of forced deleveraging in the sector. An ability to avoid daily mark-to-market, having much more control over NAV valuations and relying more on unsecured debt gives BDCs a strong leg up relative to CEFs in holding firm through market dislocations. This doesn’t mean BDCs are immune to deleveraging, particularly if, unlike 2020, we hit a “real” recession with a wave of broad-based credit losses. However, for more purely market-based temporary dislocations, BDCs should be able to withstand these better than CEFs.

Market Risk

There are different ways to define market risk with volatility being the predominant metric. We tend to avoid talking in terms of volatility because it is not particularly intuitive. And it is especially tricky for BDCs because the sector discloses NAVs only on a quarterly basis which requires us to annualize the quarterly readings to express volatility in standard terms – a process that is makes a number of assumptions as to the mathematical distribution of underlying asset prices and adds additional noise in the process.

Rather we tend to talk in terms of drawdowns and resilience. A drawdown is simply the largest peak-to-trough move within a year over some past time frame – typically 5 or 10 years. This number will tend to capture the COVID market sell-off, or the move between, roughly, the middle of February of 2020 to the third week of March since that has been the biggest intra-year peak-to-trough move over the last decade for most assets.

By this metric the loan CEF sector had a 44% price drawdown and a 30% NAV drawdown in median terms. The BDC sector, in our 25-company sample, had a median 58% price drawdown. We don’t know the NAV drawdown because of quarterly reporting but the Dec-2019 to Mar-2020 median NAV drop was around 9%. As we already discussed later this number clearly defies belief but it has an advantage for investors. The upshot here is that BDC investors should be prepared for a bumpier ride in price terms.

The other metric we look at – resilience – is the ability of an asset to climb back up after a drawdown. The intuition here is that an asset that falls 40% but gets back to even within a year is potentially more attractive than an asset that drops only 20% but “locks in” that loss and is unable to recapture its previous peak even if underlying prices rise back up (something we have seen in many mREITs).

A good way to look at this figure is simply to check full year 2020 returns. Loan CEFs delivered about a 4% total NAV return and about a flat total price return in 2020. BDCs delivered about a 4% average total NAV return and a 8% price loss in 2020.

It is interesting to note that NAV returns for the two sectors were the same over 2020, loan CEFs had much larger NAV drawdowns but BDCs had significantly larger price drawdowns. Investors who are highly sensitive to price drawdowns (perhaps, because they may lack conviction in their allocations or run leveraged portfolios) should take this into account.

Performance Dispersion

Performance dispersion within a given sector should play a part in investor allocation. For example, a sector where nearly all or most securities deliver pretty much the same return (i.e. having low return dispersion) each year should take up less investor focus and vice-versa, a sector with high return dispersion should take more investor focus as the potential return from careful intra-sector allocation has more potential upside. We calculate a measure of dispersion by averaging the annual performance differential between the 75th and 25th percentile returns in the sector i.e. the difference in annual performance between a security that bested 75% of the sector less the security that beat only 25% of the sector for that given year.

Loan CEFs have an average price dispersion of around 5% and NAV dispersion of 2.6% over the last 5 years. BDCs have price dispersion of 15% and NAV dispersion of 5.8%. However you cut it, BDCs have much larger dispersion of annual returns. This likely has to do with larger differences in leverage and asset class allocation, particularly equity holdings.

These figures have two possible interpretations. One is that BDCs offer investors greater potential alpha from careful allocation for any given year over loan CEFs since the upside for picking winners any given year is much greater in BDCs. The other interpretation is that investors who want more-or-less “typical” BDC returns need to diversify more in BDCs than in loan CEFs to ensure an average return and cut down on the element of luck (good or bad) in their portfolios.

NAV Trend

Another key consideration for investors comparing the two types of investment vehicles is the NAV trend. One comment we have seen is that BDCs follow a broken business model because they are forced to distribute at least 90% of their taxable income (and close to 100% in practice) which leaves them with little to cover credit losses, condemning them to a downward NAV trajectory and forcing them to repeatedly issue new equity, potentially diluting existing investors.

Of course, this couldbe the case as is true for many credit-focused CEF sectors (such as agencies, high-yield bonds, investment-grade bonds and others) or vehicles such as agency-focused mortgage REITs (where 4 of 5 agency-focused mortgage REITs have book values below their Dec-19 levels due to the deleveraging in the sector), however, it doesn’t appear to be the case for BDCs.

The chart below shows normalized (i.e. starting at 100) net NAV trends (excluding a couple of companies with recent splits for clarity). It is true that a few BDCs do feature NAVs that are well below their starting points 5 years ago, but just about as many are above that starting point also.

Source: Systematic Income

In our view what has allowed BDCs to maintain their NAVs are 1) the fact that they tend not to overdistribute relative to their portfolio yields (unlike most fixed-income CEFs whose income yield is above their portfolio yield), 2) a relatively mild credit environment over the past 5 years and 3) the fact that most BDCs have a small allocation to common shares or warrants which can provide additional upside. This last feature makes BDCs more attractive than credit CEFs which tend to primarily allocate to credit assets because it provides potential additional upside to NAVs during risk-on markets.

Likely Market Environment

Finally, it is important to highlight the impact of the likely market environment over the medium term and how it may impact credit CEFs and BDCs. The key feature of BDCs, in contrast to most credit CEFs, is that their holdings are primarily floating rate while their debt profile is mostly fixed-rate. On average, 91% of BDC debt assets have floating rates while 37% of debt is floating-rate.

Source: Systematic Income

In terms of interest rate sensitivity we need to take into account that there is roughly twice as many assets as debt on average so the beta to higher rates is higher than these numbers suggest. Two things to take into account, however, is that many BDC debt holdings have Libor floors between 0-3% which means they will not see an immediate jump as soon as the Fed hikes rates. Secondly, because Libor resets quarterly, there will tend to be a lag of up to a quarter even in the absence of floors. And thirdly, not all BDC investments are debt investments though the vast majority is, on average.

In contrast, the majority of credit CEFs are the other way around – they tend to have fixed-rate securities (only about 30 out of 250 or so credit CEFs have primarily floating-rate holdings). And the majority of CEFs have floating-rate credit facilities. What this means is that unlike BDCs which should benefit from rising short-term rates (with some delay given the caveats above), most credit CEFs will be hurt by rising short-term rates.

Valuation

In valuation terms, neither credit CEFs nor BDCs look amazing at the moment. The chart below captures the average and median valuations in our 25 BDC coverage. If there is good news here is that the aggregate valuation metric has not quite reached the previous peaks.

Source: Systematic Income

As far as credit CEFs, the orange line in the chart below shows that we are a bit closer to previous valuation peaks than are BDCs.

Source: Systematic Income

Obviously, there is much dispersion around these aggregate metrics and, though, neither investment vehicle aggregate valuation looks amazing, BDCs are perhaps a touch ahead here on this score.

Highlighting FDUS

There are different ways for investors to allocate to individual BDCs. A good starting point that most investors will probably agree on is to check two key metrics – longer-term total NAV returns and current price / NAV valuation which is shown in the chart below with the most attractive quadrant (high NAV returns and inexpensive valuation) highlighted in green. Although the pattern is not perfect, there is a clear relationship in the chart of worse-performing companies carrying a low valuation and vice-versa which is not a coincidence.