"It's not a question of enough, pal. It's a zero-sum game, somebody wins, somebody loses. Money itself isn't lost or made, it's simply transferred from one perception to another."
- Michael Douglas as Gordon Gekko, Wall Street
Securities that have high yields usually have high yields for a reason. Typically they entail a high degree of risk or the market is correctly discounting the probability that such yields are unsustainable (I have written about one such example, Ferrellgas Partners FGP, here). On occasion the markets will serve up a fat pitch due to a macro dislocation, but usually we have to go scrounging for opportunity. For those that are willing and able to put in the effort, complexity is our friend. What might appear fairly valued on a simple valuation or yield screen, may contain hidden treasure.
Until late 2010 I never paid attention to the municipal bond market. I'm more of an equity guy and always considered the muni market a bit of a boring backwater that I would eventually get around to looking at about 10 minutes before my retirement. That changed when the muni market went bananas in December 2010 when noted banking analyst Meredith Whitney went on 60 Minutes and stated that there would be "hundreds of billions" in losses in the municipal bond market, and that it would be "something to worry about within the next 12 months." The video can be found here.
Imagine for a moment a comfortably-retired grandpa sitting on a mountain of "safe" muni bonds settling down in front of the tube to watch Andy Rooney, like he has every Sunday since the late 1970s, and seeing this:
Although his Q3 statements would have looked fine, in fact the municipal bond market was already in the throes of a dramatic (by muni standards) 11% sell-off. The chart below shows the price (in green) and shares outstanding (in white) of the iShares National AMT-Free Muni Bond ETF (MUB). Not surprisingly, many muni bondholders panicked and liquidated holdings into a relatively illiquid market.
A muni bond trader once told me "It's easy to buy muni bonds, but it's hard to sell muni bonds." Now I know what he meant.
During that period, while running some screens on muni bond ETFs and closed-end funds, I happened upon the subject of this article: PIMCO Municipal Income Fund II (PML). At first, I was intrigued that a municipal bond fund that was trading around net asset value was able to pay out a (tax free) distribution of approximately 8%. After a deep dive on the fund I purchased shares just below $10 and (idiot that I am) sold them about a year later just over $12. Including distributions the total return was over 30%. It was painful to watch PML rise to close to $14 by the end of 2012 so I took it off my screen never to think about it again. Until now. What follows is an (updated) rehash of the work I did on PML in late 2010 that I greedily kept to myself at the time, and thankfully saved to dust off now.
The last two months have seen a savage sell-off in pretty much every interest rate sensitive security out there, including pretty much every type of bond (including muni bonds), and rate-sensitive equities such as utilities. MUB sold off over 9% in May and June before recovering somewhat:
A feature of ETFs is that they typically do not deviate much from their underlying Net Asset Value due to the ability of market participants to arbitrage any difference through the delivery or redemption of creation units of underlying securities.
Closed-End Funds ("CEFs") by contrast can and do fluctuate significantly around NAV based on the whims of Mr. Market. This can create opportunities to buy and sell at (justified or unjustified) premiums or discounts to NAV. CEFs like PML, especially those that employ leverage have been especially mauled over the past two months. Not only do these CEFs suffer from a decline in NAV (as panicked muni investors hit the sell button), but they tend to be less liquid and have had their NAV premiums evaporate. See the chart below for PML:
In the past two months PML has declined in price by approximately 17%. To equity market participants this may simply be a rough couple of months. To municipal bond investors, a 17% loss is a catastrophe.
PML's decline can be explained by an approximately 11% decline in NAV, coupled with a NAV premium that has gone from about 6% to just over 0% as shown below:
What About The Supposed Hidden Treasure?
Before getting into the nuts and bolts of PML and its underlying portfolio, I want to explain why I believe that PML's economic value is significantly greater than book NAV. The reason is not that I think the value of the underlying muni bonds is somehow mispriced, it is because of the unique and esoteric form of leverage employed by PML. Specifically, PML equity holders have the great fortune of owning shares in an ETF that uses legacy auction rate preferred securities for leverage. I will explain in more detail below, but essentially while PML's preferred shareholders technically have a claim on PML's assets, the reality is they have no way to assert that claim and receive virtually no economic participation in CEF's portfolio income. The preferred shareholders are trapped in an instrument they can't redeem, and pays them virtually nothing. I believe this effectively amounts to a permanent transfer of wealth from unfortunate auction-rate preferred holders to the funds' common equity holders, and thus a material premium to book NAV is warranted. Even if one does not share that view, the situation allows for equity holders to earn a super-normal return, particularly when measured against other CEFs or ETFs.
Auction Rate Securities?
An "auction rate security" or "ARS" is simply a debt or preferred share instrument that has a long-term maturity, but has its interest or dividend rate reset on a regular (often weekly) basis. Investors in the ARS market didn't think twice about the long-dated maturity (often multiple decades) as the market was fairly liquid, and given rates reset frequently they did not bear a lot of interest rate risk.
The ARS market grew to over $200 Billion by 2008. As complex instruments with a minimum investment size of $25,000, the bulk of ARS were/are owned by institutions and high net worth individuals. Too bad for them as you will see.
Prior to the global financial crisis, an unwritten rule of the ARS market was that auction-running broker dealers would provide a secondary market for securities between auctions, and broker-dealers almost always bid on their own auctions to prevent the risk of a failed auction.
In February 2008, virtually every auction failed. Concerned for their own liquidity more than that of their Muppets, errr, clients, the broker dealers refused to bid or make a market on a secondary basis. Some banks were cajoled by regulators to buy back securities it had sold to charities, individual investors, and small businesses. Citigroup alone had to swallow $7.3 Billion of garbage - just what it needed going into the Lehman Brothers collapse a couple of months later.
PML's auction-rate preferred shares ("ARPS") have failed since mid-February 2008.
An important "feature" (or curse, if you happen to own one) of PML's ARPS is that with failed auctions, the ARPS default to a defined "maximum rate" that is defined as follows:
"ARPS holders have continued to receive dividends at the defined "maximum rate" equal to the higher of the 30-day "AA" Composite Commercial Paper Rate multiplied by 110% or the Taxable Equivalent of the Short-Term Municipal Obligations Rate-defined as 90% of the quotient of (A) the per annum rate expressed on an interest equivalent basis equal to the Kenny S&P 30-day High Grade Index divided by (B) 1.00 minus the Marginal Tax Rate (expressed as a decimal) multiplied by 110% (which is a function of short-term interest rates and typically higher than the rate that would have otherwise been set through a successful auction)" Source: PML 2012 Annual Report
The technical term for the output of the formula above is "jack squat." For the 6-month period ended November 30, 2012 PML's 5 series of ARPS paid out a total (annualized!) rate of 0.274%. The chart below shows the 30-day US Commercial Paper index according to Bloomberg:
It's hard to see, but the rate is now down to 0.17%, which would bring the ARPS dividend down to 0.19% or so.
Holders of PML's ARPS are stuck holding an instrument that was originally viewed as a "cash equivalent" that pays them perhaps 20-30 basis points in annual yield. Thanks to our friends at the Federal Reserve who have pledged to keep short-term rates very low for a very long time, this situation is set to remain for years. ARPS holders can look longingly at rising yields further out the curve in response to the market's "taper tantrum," and enjoy years more of miniscule yield until short-term interest rates eventually rise.
Oh, did I mention these things never mature? Not only are ARPS holders left holding an illiquid instrument that pays them virtually nothing, but they never pay off! Ouch.
What would you do if you were one of the unfortunate saps who stretched for a few extra basis points of yield and owned this stuff? Flip out, of course. And flip out they did. Right across the board holders of ARS demanded that someone, anyone, make them whole. This led to pressure on banks and brokers in particular (vs. the issuers of ARS themselves) to buy back ARS at par from individuals and small businesses, such as the Citi repurchase mentioned above.
Allianz (owner of PIMCO) announced on March 24, 2008 that it was "working to address the illiquidity of ARPS." The original hope of ARPS bagholders was that funds would refinance the ARPS and redeem at par. However Allianz noted that issuing debt financing was not practical for municipal closed-end funds as the funds are not taxable - thus the cost of debt (vs. preferred equity) would prohibitively be expensive. Two months later in May 2008, Allianz put a final nail in the redemption coffin by stating that refinance options are too risky for common shareholders whose returns would decline if the cost of leverage increased (yes, some people care about fiduciary duty). The letter notes:
"A shift in the funds' financing facilities would increase the future risk of short-term financing being withdrawn or re-priced at inopportune moments, thereby damaging the interests of some shareholders"
ARPS holders had a partial reprieve when, during the depths of the financial crisis in late 2008, PML was forced to redeem a portion of ARPS to keep its leverage ratio in line (more on that later). In an ironic twist, some common shareholders were irate that such securities were redeemed at par, even though it was the only practical way of quickly reducing leverage in the fund.
In December 31, 2011 Allianz issued the following letter to the preferred shareholders:
Translation: you're out of luck.
Practically speaking there is no conceivable circumstance that cheaper financing is or will be attainable to refinance the preferred shares (and as a common shareholder if for some method is conjured up that would be fine), and the only way PML will redeem the ARPS is if they are forced to as a result of being offside from a leverage ratio perspective.
Below is a calculation I did using the daily NAV as of July 15 of $11.13 reported by Allianz/PIMCO here to determine where PML is from a leverage standpoint. The key leverage test is effectively the requirement that PML's portfolio must be 200% of the $367 million in ARPS outstanding.
As you can see, even after the recent mauling in the muni bond market, PML is running an asset coverage ratio in the 285% range, well above the minimum threshold of 200%. I calculate that the portfolio (of municipal bonds) would have to decline by almost 30% before this would occur. This is extremely unlikely for an investment-grade muni bond portfolio managed by PIMCO. I'd note that as of March 31 only 11% of the portfolio was exposed to general obligation (GO) bonds, with 84% being revenue bonds tied to hospitals, tobacco settlements, etc.
This is a long-winded way of stating that PML has locked in a very low (far below market) long-term cost of leverage. This amounts to a permanent transfer of wealth from ARPS holders to the common shareholders of PML.
PML is managed by PIMCO, the world's largest (and in my opinion one of the most sophisticated and professional) fixed income manager. The fund's stated objective is to "seek current income exempt from federal income tax." Since 2011 PML has been managed by Joe Deane, a 41-year veteran of the investment industry and head of municipal bond portfolio management for PIMCO. Joe's bio and some of his writings can be found here.
PML has paid a monthly distribution of 6.5 cents per month since January 2007, or 78 cents annual. On the current share price of approximately $11.02, this represents a current yield of approximately 7.1%. At a 35% tax equivalent basis this would roughly equate to an 11% taxable equivalent. But we must look under the hood to assess the underlying portfolio vs. just what the fund happens to pay out.
PML provides decent portfolio transparency. A high level summary as of May 31, 2013 is below:
While interesting the above portfolio details are lacking in some key details. Most notably we are not told the underlying yield-to-maturity (YTM) or yield-to-worst (YTW), and it's not clear what exactly "average coupon rate" and "average duration" are - are these weighted averages?
Fortunately on a quarterly basis PML discloses the entire portfolio. As such we know every security as of June 28, 2013. The file can be found here. Unfortunately this is pretty raw data and there are over 200 securities to analyze.
I was able to get the data into Excel and do some basic metrics. On a total portfolio value of approximately $1.05 Billion, I calculated a weighted duration of approximately 8.9 years (close to the "average" of 9.13 as of May 31 reported above), and a current yield of approximately 5.44%.
However, most of the bonds are trading above par (a weighted average of about 104 for each 100 of bonds, across various maturities), so the "current yield" overstates the actual yield to maturity as it does not account for the capital loss that will occur as the bonds ultimately mature at par (they don't own ARPS). Fortunately Bloomberg's portfolio analytics functionality is able to give us a calculated YTM and YTW. With the caveat that Bloomberg did not have data on 17 of the 200+ securities in the portfolio, the YTM is reported at 5.14% and YTW at 4.56%.
I'm going to use 5% as a SWAG estimate of the true underlying return on the portfolio going forward. This of course assumes no future alpha (outperformance) from PIMCO - it essentially assumes the assets are held to maturity and interest and principal are reinvested at the same rate. This is a more appropriate way of looking at the portfolio as PMLs' statement of operations only includes interest income in the calculation of Net Investment Income and only picks up unrealized appreciation / depreciation below the line.
Based on the above, I calculate the following:
As shown in the above rough estimates, it seems that the underlying investment return is within spitting distance (90%+) of the current distribution yield. Given the significant buffer on the Asset Coverage Ratio shown above it seems there is little risk of a cut for the foreseeable future. However, I focus less on the distribution and more on the attractiveness and economics of the underlying portfolio.
Trading at NAV today, what's striking in the above analysis is that while ARPS holders have put up about 35% of the capital for the portfolio ($367M on a $1.06B portfolio), they are only receiving about 1.7% of the return from the portfolio ($734K out of a $44M return). This is why I believe a premium to NAV (10% or greater) is justified.
This situation effectively allows taxable U.S. investors to earn a 6-7% tax-free return, which is equivalent to approximately 10% on a pre-tax basis. On a portfolio of investment-grade muni bonds. Once the dust settles in the fixed income market it's not too difficult to imagine PML moving back from $11 currently to the $13 range that it was at about 8 weeks ago which would still offer a very attractive 5.5% after-tax yield, and nearly 20% upside from current levels.
One of the risks to the portfolio is that, sometime (perhaps 2015 or later) short-term interest rates could start to rise. However even with the ARPS rates moving up 200 basis points (from 0.2% currently) to 2.2%, it would only knock about 1% off the economic yield. The bigger risk perhaps is in owning a relatively long-duration (9 year) fixed income portfolio in the event of a large rate increase. For my views on why I am not worried about a massive interest rate increase please look here. The portfolio is actively managed by PIMCO so I would not expect them to stand still if market conditions change. I would also note that the muni yield curve is pretty steep already and investors are being well compensated for the risk, especially after the recent sell-off in fixed income markets (particularly after tax!):
Finally, for those Bill Gross fans out there, I would point out that he (and his family) owns 554,460 shares of PML worth over $6 million, after selling a relatively small amount (72,000 shares) in June at a price of $11.57 out of two child trusts.
In 1973 two Stanford economists (Edward Shaw and Ronald McKinnon) coined the term "financial repression" to refer to governments that are able to channel funds to themselves using techniques such as capping interest rates paid to others. PML is a rare opportunity to implement your very own flavor of financial repression by sticking it to the ARPS holders - a captive supply of permanent dirt-cheap capital.
"There are idiots who will buy anything as long as it costs enough."
- Andy Rooney
Additional disclosure: This article reflects my personal views only. I have a long position in PML. All data and calculations presented are accurate to the best of my knowledge but have not been vetted, checked, proofread, or independently verified. This article does not constitute investment or tax advice, and should not be relied upon for any purpose other than entertainment. I welcome comments and or corrections.