Yield-starved investors have ventured into exotic – and often risky – assets, including hedge funds, non-traded REITs and private placements. But an asset class that has been around since 1893 offers a compelling combination of low risk and high income. A carefully selected portfolio of closed-end funds (CEFs) will yield 8% with less volatility than the S&P 500.
I’m not alone in paying renewed attention to this asset class. In late 2011, Burton Malkiel wrote an op-ed in the Wall Street Journal urging investors to replace traditional bonds in their portfolios with, among other things, leveraged municipal bond CEFs. Malkiel’s suggestion was especially notable given that he is known as a champion of simple portfolios constructed from low-cost market-cap weighted open-end mutual funds, an investment philosophy he most famously espoused in his classic book, A Random Walk Down Wall Street.
In this article, I will explore the general characteristics of CEFs and how to identify those that are the most attractive for income investors. While CEFs have additional complexity relative to open-end mutual funds, the best provide investors with very attractive income on a risk-adjusted basis. Lipper estimates, for example, that the average CEF is yielding 7.3%, although that claim should not be taken at face value (more on this later). Given the paltry yields in other parts of the capital markets, advisors and investors need to consider an allocation to these funds.
What is a CEF?
CEFs have many similarities to the open-end brethren. Investors can buy or sell shares of a CEF like a traditional open-end mutual fund. The fund manager charges fees against the assets in the fund for managing the fund’s holdings. Like an open-end fund, CEF shares are traded on an exchange, rather than through the mutual fund company that manages the fund.
But with an open-end mutual fund the manager creates or redeems shares to meet the balance of supply and demand. The key distinction of a CEF is that it has a fixed number of shares, which can be altered only if the manager does a secondary offering. The price of CEF shares often drift away from the net asset value of the fund holdings, trading at either a discount (less than the value of the fund’s assets) or at a premium (more than the value of the fund’s assets).
Morningstar has a useful resource center with a range of information about the basic attributes of CEFs. Readers who are unfamiliar with this fund structure will benefit from going through Morningstar’s slideshow that explains the basic attributes of CEFs. The Closed-End Fund Association (CEFA) is a trade group that provides a range of resources for researching these funds.
The tendency of CEF shares to trade at values that are quite different from the NAV of their holdings represents a challenge to financial theory. Burton Malkiel penned a 2005 article in which he referred to the persistence of differences between the prices of CEFs and their NAVs as one of the most enduring conundrums in the field of finance. This anomaly appears to violate the law of one price: How is it possible that a basket of securities (a share of the CEF) can trade at a substantially different price than do the assets in the basket?
Aside from the CEF discount or premium, investors need to understand other key characteristics of a CEF. There are different forms of income that CEFs generate: income derived from the underlying assets (so-called income-only yield) and the distribution yield or market yield, which may include dividend or coupon payments, short- and long-term capital gains, and return of capital. Income-only yield, if there is any income, is a component of the market yield. When the market yield is greater than income-only yield, it is often the case that the additional distribution to investors is simply return of capital.
Given the complexity of CEFs, are they worth the bother? There are several potential benefits to the CEF structure. First, the fund’s ability to use leverage allows more latitude in managing the portfolio. Because the fund company itself cannot be forced to redeem shares, CEFs are an excellent vehicle for portfolio of relatively illiquid assets. CEF managers do not have to maintain a balance of cash to meet potential investor redemptions. Nor do they need to be prepared to deploy large inflows of new cash (as open-end fund managers do).
CEFs are more flexible as an income-generating fund structure for these reasons. Indeed, in a survey of investors, Morningstar found that income generation was the primary motivation for choosing CEFs.
Finding the best CEFs
Analyzing CEFs is complicated because of the discount or premium of share price to NAV, the different forms of distribution (such as income or return of capital), and the variability in leverage between funds. I have created a simple approach to respond to those concerns. To judge the relative risk and yield of these funds, I looked only at the component of distributions that is attributable to traditional income streams (stock dividends and bond coupons). To this end, I based my analysis on the income-only yield data for CEFs compiled and tracked by CEFA. This allows us to look at the yield-versus-risk of a CEF to provide a baseline comparison to other income-generating asset classes.
One of the challenges in assessing CEF yields is how they are expressed. Morningstar, for example, annualizes the most recent distribution (as opposed to looking back over the last 12 months) and then calculates distribution yield from this annualized value and the current price of the CEF. CEFA, for its income-only yield, uses net income figures from the latest annual report divided by the average net assets in the fund. Both of these are reasonable approaches, but they may differ substantively if the income from a CEF varies significantly from year to year, or through the calendar year. The goal of both methods is to create a consistent projection of income, but there may be circumstances in which the most recent year’s income is not a good predictor of the next year’s, or there may be those in which annualizing the most recent income payment is a poor predictor of the entire year. I have created an additional check of both approaches simply by looking at the income distribution for 2011 (the most recent full calendar year) and the current price.
I selected those CEFs with (a) the highest values of CEFA’s income-only yield and (b) at least three years of history. The income-only yield provided by CEFA is based on net asset value, rather than market price of the fund shares, so I adjusted to the yields based on current market prices to create a yield that is the equivalent to the way that yields are expressed for other assets. I also subtracted the expense ratio of the fund from the yield.
I ran a Monte Carlo simulation to generate risk projections for each fund, ranking the CEFs on the basis of the ratio of yield-to-risk, using the net income-only yield, minus expenses. Among the 20 CEFs with the highest yield-to-risk ratio, I computed an income yield based on market price using the recent closing price of each CEF and the income portion of the total distribution from Morningstar. I was principally interested in identifying funds that seemed to have substantially lower incomes for the YTD or the most recent distributions than they had previously, because that indicates the fund is less likely to generate consistent income.
The volatility projections I used in my simulations have been extensively tested for a range of asset classes and are very close to the trailing three-year historical volatilities.
The 20 CEFs with the highest estimated yield-to-risk ratio are shown below. This table also shows the projected volatility for each of these funds, as well as the estimated income-only yield relative to market price. The latter is derived from CEFA’s income-only yield and my own estimate of income-only yield, for which I used 2011 income and the current price for each CEF.
Top-Ranked CEFs on the Basis of Yield vs. Risk (Net of Expenses)
These funds have the highest calculated income-only yield relative to their risk levels. For reference purposes, the projected volatility of the S&P500 is 22.5%, based on the implied volatility of long-dated options on SPY.
The average income yield of the CEFs in the table above is 8% (from the CEFA-based numbers), or perhaps 8.3%, based on my estimates using 2011 income. The average volatility for these funds is 17.7%. These results indicate that it is possible to generate income yields of 8% or more with substantially less volatility than that of the S&P 500. An equal-weighted portfolio of all 20 funds has a yield of 8% with a projected volatility of 15.1%, indicating a benefit from diversification. This portfolio has beta and R-squared (with respect to the S&P 500) of 0.48 and 53%, respectively, which shows that about half of the variance in return is driven by the systematic risk of the equity market.
To see how these CEFs compare to the broader population, the chart below compares them to the yield and risk for the broader universe of all CEFs analyzed and a high-yield bond index ETF (ticker: HYG)
CEFs with highest yield vs. risk (blue circles) vs. 70 CEFs with highest income-only yield from CEFs (red triangles) and high-yield bond index fund (HYG, green square)
Several features of this chart are striking. First, those 20 CEFs are more attractive than most other CEFs, even though all of the funds were selected on the basis of high income-only yield. Potential investors need to look well beyond raw yield numbers in their search for an appropriate fund.
The yields and risk for the top 20 funds have a roughly linear relationship, and the funds with the lowest yields and lowest risk levels are comparable to high-yield bonds. Conceptually, this makes sense. The high-yield bond ETF is un-levered, and could be leveraged to increase yield but only with a commensurate increase in risk.
Conclusions
While unfamiliar to many advisors and investors, CEFs offer attractive values in the current environment. Their yield-versus-risk profile positions them as a viable alternative asset class in investors’ long-term portfolios, especially when compared to strategies such as hedge-fund replication.
When evaluating CEFs on the basis of yield-versus-risk, I adjusted for the fact that not all of the distributions are equivalent to income. The process of estimating income-only yield is, however, straightforward. The CEFs with the highest income net of expenses relative to their risk levels are the most attractive. The highest-yielding among these have income levels around 10% per year, with risk levels close to those of the S&P 500.
An equal-weighted portfolio of the 20 CEFs with the highest income yield-to-risk ratio would have an income-only yield of 8% and a projected volatility of 15.1%, roughly equivalent to the statistics for a portfolio that is 70% allocated to the S&P 500 and 30% allocated to an aggregate bond index. But a 70/30 portfolio has yield that is just above 2%. The 8% yield from the equal-weight CEF portfolio is very attractive, even in the context of total return. A 70/30 portfolio, to put it bluntly, does not have an expected total return of 8%.
A recent paper argues that constraints on the use of leverage led to the mispricing of assets, such as CEFs. Given the high yield–versus-risk for the more attractive CEFs, the widespread use of leverage may be allowing these funds to exploit this arbitrage opportunity. According to Morningstar, more than 70% of CEFs employ leverage. Particularly in the current low interest rate environment, CEFs can boost income because leverage is so cheap.
An 8% yield for the same volatility as that of a 70/30 portfolio may seem too good to be true. Are there additional risks that are not captured in the historical and projected volatility, thereby justifying the high yields? Given that there is no consensus opinion that explains CEF premiums and discounts, and that this effect is a true anomaly in finance, it is certainly possible that there are some special risks that our volatility projections may be missing.
One potential source of risk is the ability of a CEF to maintain its distributions. While there are long-term studies of the consistency of dividends from stock and coupon payments from corporate bonds (both investment grade and junk), research is lacking on this topic for CEFs. My cursory examination of the distribution history of CEFs makes it clear that distributions can vary greatly from year to year. But in theory, of course, such variability should be reflected in the volatility of the CEFs’ prices.
Investors challenged by the current low-yield environment will be well served by allocating a portion of their portfolios to carefully chosen CEFs. While a foray into this class of funds requires familiarity with some new terminology and features, this is an investment of time and effort that is well worth making.
Geoff Considine is founder of Quantext and the developer of Quantext Portfolio Planner, a portfolio management tool. More information is available at www.quantext.com.
Geoff’s firm, Quantext is a strategic adviser to FOLIOfn,Inc. (www.foliofn.com), an innovative brokerage firm specializing in offering and trading portfolios for advisors and individual investors.
Read more articles by Geoff Considine
