UBS Puerto Rico's Bond Fund Debacle: What We Know so Far

Over the past year, we've posted a dozen short commentaries to our blog post here. We thought it would be helpful to summarize what we know so far. This summary and our prior UBS Puerto Rico blog posts are available in Spanish. You can find all the Spanish-language blog posts by clicking here.

We discussed the national exposure of the UBS Puerto Rico losses in October 2013 in Trouble in Paradise: UBS Puerto Rico Bond Fund Investors Hit Hard. The losses only got worse thereafter. In calendar year 2013, a series of municipal bond closed-end Funds managed by UBS Asset Managers of Puerto Rico and sold by UBS Puerto Rico brokers declined by up to 60% in price.

Table 1 shows the NAV and Capital Losses by Fund. The NAV total returns for 2013 ranged from -22.2% to -58.7%. The approximate dollar losses across the 19 funds were $2.1 billion based on the NAVs and $2.85 billion based on bid prices.¹




_______________________________________

¹This calculation uses a constant value for the shares outstanding obtained from each of the Funds' annual report closest to December 31, 2012. The Funds have different fiscal year end dates and hence the calculation is an approximation.

The average US mainland municipal bond fund only fell about 2.5% in 2013. Figure 1 plots the value over time of six large UBS Puerto Rico Funds along with the value of the S&P Municipal Bond Puerto Rican Index, Vanguard 500 Index Fund and Vanguard Total Bond Market Index Fund, all standardized to $10 on December 31, 2012. While Puerto Rican municipal bonds lost 20% the UBS Funds lost 50% to 60%.




In "Diversification and UBS Puerto Rico Bond Fund Losses" we show that the UBS funds were concentrated in the bonds of two issuers in 2013. For most of the Funds, the securities of either the Employee Retirement System (ERS) or of the Sales Tax Financing Corporation (COFINA) accounted for over 25% of the fund's gross assets - and over 40% of the Fund's gross assets for the two issuers combined. We also showed that the allocation of the $1.7 billion in losses investors in the UBS Puerto Rico Funds suffered between December 31, 2012 and September 30, 2103 were directly related to how much UBS concentrated the Funds into the bonds of these two issuers.

Table 2 shows the concentrated holdings in the two issuers in the 19 UBS PR funds which were not in the distribution phase as of December 2013. The table also shows the losses incurred by each of the Funds. Funds with higher concentrations on ERS and COFINA bonds suffered greater losses.




In The Use of Leverage in the UBS Puerto Rico Closed-End Funds Magnified Losses we show that losses in the value of Puerto Rican bonds in 2013 were magnified in the UBS Puerto Rico Funds by UBS's high use of leverage.

Mainland closed-end funds are generally limited to borrowing no more than 33% of their gross assets unless they issue preferred stock in which case they can borrow up to 50% of their gross assets. Closed-end funds can leverage more if they issue preferred stock because doing so exposes fund investors to less risk than if the funds use short term borrowing. The UBS Puerto Rico Funds could - and did - borrow 50% of their gross assets without issuing preferred stock so investors who bought UBS Puerto Rico Fund shares had a 2-1 leveraged exposure to the Fund's holdings.

Table 3 lists the UBS Funds' borrowing as a percent of gross assets on June 30, 2013. The borrowing ranged between 44.01% and 50.74% of gross assets - right at the 50% maximum amount allowed by Puerto Rican securities regulations. Figure 2 shows that the UBS Puerto Rico Funds used leverage much more aggressively than other closed-end municipal bond funds.







In addition, some media reports have also pointed out that investors may have been encouraged to borrow money to invest in the UBS Funds. This leverage, external to the funds, would further magnify investor losses.

The Concentrated Positions Were Built Exclusively in UBS-underwritten Bonds

The concentration in the UBS Funds was built exclusively in bonds that were underwritten by UBS and bought by UBS into the Funds' portfolios at prices retail investors typically pay in the offering. UBS's purchases of entire issues it underwrote into the Funds at the initial offering created a significant conflict of interest between UBS as underwriter and UBS as fund manager with a fiduciary obligation to fund investors. UBS, as an underwriter, received underwriting fees which in part compensated UBS for selling effort which UBS minimized by selling directly into its proprietary funds. In addition, UBS's management and marketing of the Funds allowed UBS to underwrite more bond issues than it otherwise could without the funds as repositories for the large bond positions that UBS underwrote. UBS also minimized fund managerial effort by relying on in-house underwritten bonds rather than researching and purchasing appropriate bonds in the secondary market or from other underwriters.²

In UBS Stuffed $2.5 Billion of ERS and COFINA Bonds it Underwrote in Its Puerto Rican Funds in 2007 and 2008 we illustrated how UBS dramatically changed the UBS Puerto Rico Funds in late 2007 and early 2008 using the Tax Free Puerto Rico Fund II and Puerto Rico Fixed Income Fund II as examples. UBS sold off $350 million other bonds and bought $530 million of ERS and COFINA bonds it underwrote in these two funds alone.

Table 4 summarizes the market value of Tax Free Puerto Rico Fund II holdings by issuer on June 30, 2007 (from 2nd Quarter 2007 Quarterly Review available here), November 30, 2007 (from the Annual Report available here) and on November 30, 2008 (from the Annual Report available here).

The Tax Free Puerto Rico Fund II held no COFINA bonds on June 30, 2007. UBS participated in COFINA underwritings on July 13, 2007 (COFINA Series 2007A Offering Circular available here) and on July 18, 2007 (COFINA Series 2007B Offering Circular available here). The Tax Free Puerto Rico Fund II Fund owned $20.3 million of COFINA bonds on November 30, 2007. UBS was the sole underwriter of the (COFINA Series 2008A Offering Circular available here.)

The Tax Free Puerto Rico Fund II held no ERS bonds on November 30, 2007. UBS participated in underwriting three series of ERS bonds in 2008. The Series 2008A Offering Circular is available here, the Series 2008B Offering Circular is available here and the Series 2008C Offering Circular is available here.

__________________________________
²Concentrated portfolios in general (regardless of whether they are concentrated in affiliate-underwritten securities) require less managerial effort because fewer issuers must be researched and followed by the fund manager.




Between June 30, 2007 and November 30, 2008 UBS caused the Tax Free Puerto Rico Fund II to pay $188,943,300 for the ERS and COFINA bonds and these two issuers' bonds went from 0% of the Tax Free Puerto Rico Fund II portfolio to 50% of the gross assets and 98% of the net assets of the Fund. UBS sold roughly $190 million of other issuers' bonds out of the Tax Free Puerto Rico Fund II to make room for the ERS and COFINA bonds it was underwriting.

Table 5 presents holdings by issuer for the Puerto Rico Fixed Income Fund II on June 30, 2007 (from Quarterly Review available here), November 30, 2007 (from the Annual Report available here) and on November 30, 2008 (from the Annual Report available here).

The Puerto Rico Fixed Income Fund II held no COFINA or ERS bonds on June 30, 2007. Between June 30, 2007 and November 30, 2008 UBS caused the Puerto Rico Fixed Income Fund II to pay $340,357,374 for the ERS and COFINA bonds and these two issuers' bonds the ERS and COFINA bonds went from 0% of the Puerto Rico Fixed Income Fund II portfolio to 44% of the gross assets and 88% of the net assets of the Fund. UBS sold roughly $335 million of other issuers' bonds out of the Puerto Rico Fixed Income Fund II to make room for the ERS and COFINA bonds it was underwriting.




UBS Compromised Its Investors Underwriting and Buying Unmarketable Bonds

In UBS Succumbed to Conflicts and Purchased $1.7 Billion of Employee Retirement System Bonds into its Puerto Rican Municipal bond Funds in 2008 we showed that the dramatic change in the UBS Funds documented above occurred because UBS underwrote unmarketable ERS bonds in 2008 and had to buy them into the UBS Funds or lose out on tens of millions of dollars of underwriting fees. These conflicted ERS bonds turned out to be the source of most of the losses in the UBS Funds in 2013.

The Puerto Rican Employee Retirement System was acutely and chronically underfunded. Figure 3 is a plot of the PR funding ratio and the median of the 50 states' funding ratios. The states' median funding ratio fluctuated between 80% and 100% from 1990 to 2013 while the PR funding ratio was approximately 20% until 2008 after which it dropped even further. For context, the next three worst average funding ratios from 2007 to 2011 were Illinois at 51%, Connecticut at 58% and Kentucky at 59%.




Merrill Lynch attempted to sell $7 billion of ERS Pension Obligation Bonds in 2007 but failed. Near contemporaneous analysis of Merrill Lynch's failed attempt to underwrite ERS bonds and the harm to ERS caused UBS's subsequent involvement can be found in Conway MacKenzie, Inc., October 2010, Review of the Events and Decisions That Have Led to the Current Financial Crisis of the Employees Retirement System of the Government of Puerto Rico available here.

UBS replaced Merrill Lynch as advisor to the ERS and was the lead underwriter of the ERS $1,588,810,800 2008 Series A bonds. The 2008 Series A Offering Circular is available here. We don't currently have information on how much of the 2008 Series A ERS bonds UBS committed to sell but we do know that UBS purchased 41% of the issue into its Funds. We also know that UBS bought some 2008 Series A bonds into individual Puerto Rican customer accounts.




The first page of the 2008 Series A ERS Offering Circular includes the following language.

The System currently contemplates offering additional parity Bonds (the "Series B Bonds") in other jurisdictions. The Series B Bonds would be offered by means of one or more separate Official Statements and may not under any circumstances be purchased by residents of Puerto Rico.

The 2008 Series A Offering Circular language was unambiguous protection for Puerto Rican investors in the 2008 Series A bonds since the Series B bonds, which would be on par with the Series A bonds, would only be issued if the Series B bonds passed the market test.

The Offering Circular for the $1,058,634,613 2008 ERS Series B was published on May 28, 2008 and is available here). Something dramatic happened between January 29, 2008 and May 28, 2008. Rather than only being available to non-Puerto Rican residents as promised four months earlier, the 2008 Series B ERS bonds would only be sold to Puerto Rican residents. See Figure 4.




Not only was there no market outside of Puerto Rico for the 2008 Series B ERS bonds on the terms they were being offered, there apparently was no market for the bonds in Puerto Rico either. UBS bought 89% of the 2008 Series B ERS bonds into its proprietary funds. See Table 7.




There was a third and final ERS bond series issued in 2008 - the Series C bonds. UBS was the lead underwriter of the $300 million offering in June 2008. The ERS 2008C Offering Circular is available here. As with the 2008 Series A and Series B ERS bonds, the 2008 Series C bonds could only be purchased by Puerto Rican residents. UBS purchased 38% of the 2008 Series C issue into its proprietary bond funds.

The most interesting non-public documents in the UBS Puerto Rico municipal bond saga will likely be the internal emails and memos at UBS and the communications between UBS and the ERS as they realize that there was no market for the 2008 Series B ERS bonds and the 2008 Series A ERS promise that the Series B bonds will not be sold on the island had to be abandoned.

But It Gets More Interesting... In 15 Days in Puerto Rico Cost UBS Clients Over $1 Billion we point out that even a little earlier in 2008, UBS knew the jig was up. A Moody's January 4, 2008 Ratings Report available here and Standard & Poor's January 14, 2008 Ratings Report here demonstrate that at least as late as January 14, 2008 ERS and UBS were intending to issue $4 billion of ERS POBs in the two weeks left in January 2008 and another $3 billion later in the year.

Whatever happened in the fifteen days between January 14, 2008 and January 29, 2008 it was dramatic. ERS and UBS realized they could only sell $1 billion - not $4 billion - of the ERS bonds on the terms they were being offered in January 2008. $650 million of the $1.6 billion Series A bonds were bought by the UBS Funds so there was a market for at most $1 billion. It must have become clear no one else would buy the bonds and UBS was about to lose out on lucrative underwriting fees.

COFINA 2007 Series A and Series B

In UBS PR Funds Also Bought $1.35 Billion of UBS Underwritten COFINA Bonds in 2007 and 2008 we show similar conflicts which led UBS to underwrite unmarketable 2008 ERS bonds and stuff them into the Funds led UBS to underwrite unmarketable 2008 Puerto Rico Sales Tax Financing Corporation (COFINA) bonds and stuff them into the Funds. COFINA issued two series of bonds in 2007 and one Series in 2008.

UBS Investment Bank was 1 of 19 co-underwriters of the $2.7 billion Series A bonds targeting mainland investors. The COFINA 2007 Series A Offering Circular is available here. UBS Financial Services of Puerto Rico was the lead underwriter (with 11 co-underwriters) of the $1.33 billion 2007 Series B bonds sold only to Puerto Rican investors. The COFINA 2007 Series B Offering Circular is available here.

We don't know how much of the COFINA 2007 Series B bonds UBS committed to sell but we do know UBS purchased $614 million or 46% of the $1.33 billion COFINA 2007 Series B bonds into its proprietary funds. See Table 8.




COFINA 2008 Series A

In 2008, UBS Financial Services of Puerto Rico was the sole underwriter of the $737 million Series B bonds marketed only to Puerto Rican investors. None of the other 18 underwriters of the 2007 Series A or of the other 11 underwriters of the 2007 Series B COFINA bonds participated in the 2008 Series B COFINA bonds. The COFINA 2008 Series A Offering Circular is available here. UBS purchased 100% of the $737 million COFINA 2008 Series A bonds into its proprietary funds. See Table 9.




As with the ERS bonds, UBS bought the COFINA bonds it underwrote in 2007 and 2008, making room for these bonds by selling other portfolio holdings. There was no other market for the COFINA 2008 Series A bonds or UBS used its control of its proprietary funds to maximize its fees from underwriting and selling the 2008 COFINA Series A bonds.

UBS Charged Its Puerto Rico Bond Funds Excessive Markups

In Did UBS Charge its Proprietary Puerto Rico Bond Funds Excessive Markups? we identified preliminary evidence that UBS may have charged its UBS Puerto Rico Funds excessive markups.

Oppenheimer Rochester VA Fund (ORVAX) purchased $1,645,000 of a Puerto Rican municipal bond (CUSIP 74527EFA1) in October 2011 at a full spread of $0.404 or half-spread of $0.202 per $100 face value. This markup was consistent with the published literature including our own research which documents larger markups on smaller trades.

The total spread on the Oppenheimer purchase of a Puerto Rican municipal bond was $6,645.80 and the half-spread was $3,322.90. The published literature shows municipal bond markups decline dramatically in percentage terms as trade size increases because median markups flatten out at around $6,000 regardless of trade size.

Unlike mainland bond funds, the UBS Puerto Rico Funds don't report holdings quarterly but we can tell that the UBS Puerto Rico Fixed Income Fund IV purchased $7,500,000 par amount of the ERS CUSIP 29216MBA7 between March 31, 2010 and March 31, 2011. The MSRB EMMA data reflect some large trades at very low markups but the reported trades which likely reflect the Fixed Income Fund IV purchases in this CUSIP were at a $1.25 markup.

The total spread on the UBS Fixed Income Fund IV $7,500,000 purchase appears to have been $93,750 and the half spread $46,875. On just this one trade, UBS appears to have charged its Fund's investors $40,000 to $80,000 in excessive markups.

UBS Asset Managers of Puerto Rico (UBS-AM) objected when we first identified evidence that UBS was charging its UBS Puerto Rico Fund investors excessive markups. UBS argued that since we couldn't identify which trades in the EMMA data were the UBS bond fund trades and UBS wasn't providing the data, we shouldn't speculate about whether UBS charged excessive markups or not. We received a threatening letter from UBS-AM's attorneys on December 24, 2013 which we posted in Merry Christmas from UBS Asset Managers of Puerto Rico.

It turns out, as we showed in Did UBS Charge its Proprietary Puerto Rico Bond Funds Excessive Markups? Part II that we can identify at least some of UBS's purchases of Puerto Rican municipal bonds for the UBS Funds and those purchases tell an interesting story.

Table 10 shows UBS PR Funds' $86.95 million holdings face value of the 6.45% coupon, long term Puerto Rican municipal bond (CUSIP: 29216MAN0) issued in January 2008 by the Employees Retirement System.




Table 11 lists the trading activity in this bond and it shows that the UBS Funds were the only purchasers of this $87 million bond issue and that the funds paid the $100 offering price.




There are many other examples in which the UBS Funds bought all, or virtually all, of a UBS underwritten bond from an issuer (ERS) whose bonds the mainland funds wouldn't touch. This raises a number of issues which we address in later posts. For today, we will just address the markup issue.

Our example bond was part of the 2008 A Employees Retirement System deal. The Official Statement can be found here. The underwriters paid $98.95 on average for the bonds. UBS resold these bonds, which it paid the Employee Retirement System approximately $86,033,050 in the when-issued market, to its proprietary mutual funds for $86,950,000. UBS charged its mutual fund investors a $916,950 markup over the price UBS paid the issuer in what was, economically at least, a riskless principal trade. This was a 1.05% markup on an $86 million institutional purchase.

Breen, Hollifield, and Schurhoff (2006) find the average underwriting spread on municipal bonds is 0.8% and that more than half of this spread is provided to the brokerage firm as a sales credit or gross commission to motivate the sales force. They also find that a significant fraction of large trades are done below the reoffering price at the time of the offering. There were no retail brokers to compensate for selling this bond, yet UBS charged its mutual fund investors an additional $616,950 over the average $300,000 underwriter spread on this CUSIP which would have been roughly the average non-sales credit component of the spread. Thus UBS paid the Puerto Rican Employee Retirement System hundreds of thousands of dollars less than it should have or UBS caused its mutual fund investors to pay hundreds of thousands of dollars too much - or both.

UBS Succumbed to Self-Interest and Fund Investors Suffered the Consequences

In What Hell Hath UBS Puerto Rico Wrought we demonstrated that the disastrous losses suffered by investors in the UBS PR Funds in 2013 are directly traceable to UBS putting its interests ahead of its clients underwriting and buying ERS and COFINA bonds into the Funds in 2008. Consider three examples.

Fixed Income Fund II

Table 12 categorizes the Fixed Income Fund II's November 30, 2012 holdings into ERS, COFINA and Other. ERS bonds are 27.5% of the portfolio, COFINA bonds are 21.3% of the portfolio and Other Investments are 51.1% of the portfolio. The ERS bonds lost 50.3% of their value from November 30, 2012 to December 13, 2013. The COFINA bonds lost 29.7% from November 30, 2012 to December 13, 2013. Other investments in the Fund lost between 4.8% and 15.8% of their value between November 30, 2012 and December 31, 2013. We estimated a range of possible losses on the Funds' portfolios because the Funds do not produce financial statements for the periods over which we have spanning holdings data and the number of units of each fund changes significantly over time. If the ERS and COFINA bonds had only suffered the losses suffered on the rest of the Fixed Income Fund II's portfolio, the Fund would have only lost $140.2 million instead of between $201 and $251 million.



Fixed Income Fund III

The same pattern found in the Fixed Income Fund III can be observed in the other UBS PR funds. Table 13 presents similar analysis for Fixed Income Fund III. ERS bonds are 28.1% of the portfolio, COFINA bonds are 15.1% of the portfolio and Other Investments are 56.8% of the portfolio.

The ERS bonds lost 45.4% from June 30, 2013 to December 13, 2013. The COFINA bonds lost 26.2% from June 30, 2013 to December 13, 2013. Other investments in the Fund lost between 2.6% and 12.5% of their value between June 30, 2013 and December 31, 2013. If the ERS and COFINA bonds had only suffered the losses suffered on the rest of the Fund's portfolio the Fund would have only lost $103.6 million instead of between $151 million and $197 million.



Fixed Income Fund IV

Table 14 presents similar analysis for Fixed Income Fund IV. ERS bonds are 25.2% of the portfolio, COFINA bonds are 19.4% of the portfolio and Other Investments are 55.4% of the portfolio.

The ERS bonds lost 44.8% from March 31, 2013 to December 13, 2013. The COFINA bonds lost 27.7% from March 31, 2013 to December 13, 2013. Other investments in the Fund lost between 8.0% and 12.2% of their value between March 31, 2013 and December 31, 2013. If the ERS and COFINA bonds had only suffered the losses suffered on the rest of the Fund's portfolio the Fund would have only lost $107 million instead of between $185 million and $206 million.



We estimate these three funds lost between $537 million and $654 million. The UBS underwritten ERS and COFINA bonds lost $464 million of the $537 million to $654 million in losses. That is, between 71% and 86% of the billions of dollars the UBS PR Funds lost in 2013 was the direct result of the UBS underwritten ERS and COFINA bonds for which there was no market.

Taxes

In Taxes, Puerto Rico Municipal Bonds and the UBS Funds we show that preferential tax treatment for Puerto Rican investments cannot justify what would otherwise be unsuitable concentrations of investors' portfolios in Puerto Rico municipal bonds.

Puerto Rican residents don't pay federal income tax but do pay very high income taxes. The Puerto Rican maximum marginal income tax rate is 33%, reached at only $50,000 per year of taxable income. The income on Puerto Rican municipal bonds is exempt from the state income tax that would be paid by Puerto Rican investors who bought municipal bonds issued in other states. This difference in tax treatment is one of the justifications offered for the alarming concentration of UBS clients in UBS proprietary closed end funds stuffed with Employee Retirement System bonds underwritten by UBS.

This tax justification for concentrating Puerto Rican investors in Puerto Rican municipal bonds is economically identical to arguing that investors ought to concentrate their portfolio in the single stock with the highest expected return. For example, it is equivalent to arguing that investors should put 100% of their investments (or to take the UBS approach, substantially more than 100% of their investments) in Zoom.inc because it has the highest beta and therefore the highest expected return. You don't have to be old enough to remember the tech wreck to know this is unsound advice because it ignores the very high risk associated with a concentrated and leveraged investment in Zoom.inc.

We plot the annualized return and risk from 2003 to 2012 derived from Standard and Poor's Municipal Bond Indexes in Figure 5. The red dots represent combinations of risk and return for state specific indexes, the blue dots represent national portfolios and the green dot is Puerto Rico. Notice that Puerto Rican municipal bonds had lower returns and higher risk than the mainland states and national municipal portfolios before the losses in 2013.



The slope of a line from the origin to a point on the graph is the ratio of realized returns to realized risk. This is sometimes referred to as a Sharpe Ratio. Portfolios with higher returns per unit of risk (i.e. steeper slopes) are preferred.

Ignoring taxes, Puerto Rican municipal bonds experienced lower risk-adjusted returns than the 26 individual states with 10 years of data. Contrary to parties defending the concentration in Puerto Rican investors' accounts, that judgment still holds when we incorporate taxes.

We plot the returns and risk from Figure 5 after reducing the returns by the maximum marginal income tax rate in Figure 6. The cloud of red and purple dots have become triangles and shifted in toward the origin but remain above the dotted line connecting the origin with Puerto Rico's realized risk and return.



We could complicate this analysis further by separating out coupon interest from capital gains and losses but the result remains unchanged. The tax benefit Puerto Rican investors receive increases the after tax returns and risk of investing in Puerto Rican municipal bonds relative to investing in the municipal bonds issued in other states - and the Puerto Rican municipal bonds were already much more risky than bonds from other states.

Differences in estate taxes also don't explain the concentration observed in Puerto Rican investors' accounts. Exemptions from estate taxes available to Puerto Rican investors apply to other assets - not just municipal bonds and funds that hold them - and so don't justify such concentrations in the UBS Funds and UBS underwritten bonds.

1. Exemptions from estate taxes available to Puerto Rican investors apply to other assets - not just municipal bonds and funds that hold them - and so don't justify such concentrations in the UBS Funds and UBS underwritten bonds.

2. Other estate planning techniques allow diversified portfolios to pass from one generation to the next in Puerto Rico.

3. Portfolios were concentrated in UBS Funds and UBS underwritten bonds without regard to whether the estates are small or large.

4. Portfolios were concentrated in UBS Funds and UBS underwritten bonds without regard to the age of the accountholder.

To sum up, taxes are painful but they don't justify concentrating Puerto Rican investors' portfolios in the UBS Funds or the Puerto Rican municipal bonds more generally any more than high betas justified concentrating portfolios in tech stocks.

In UBS Puerto Rican Funds Did Not Belong in Puerto Rican Investors' Portfolios we demonstrate that Puerto Rican investors should not have had their portfolios concentrated in the UBS Funds.

Pre-Tax Efficient Frontier

In Figure 7, we plot the annualized total returns to diversified portfolios of stocks and bonds ranging from 100% bonds to 100% stocks. Total returns are calculated using monthly returns from January 2003 through December 2012. We use the Vanguard S&P 500 Index fund (VFINX), and the Vanguard Total Bond Market Fund (VBMFX) to reflect the pre-tax returns investors earned during this 10 year period in taxable investments.



We also plot the risk and return of the unleveraged S&P Puerto Rican municipal bond index and the risk and return of the prototype UBS Puerto Rico closed end fund in Figure 7. There is not enough trading in the UBS PR funds and the prices UBS set on the funds was unchanged for weeks at a time so we cannot estimate the true risk of the UBS Puerto Rican Funds. The UBS closed end funds were leveraged 2:1. So if the Funds' portfolios matched the S&P Puerto Rican Index they would have been twice risky as the S&P PR Index. In fact, the UBS Puerto Rican Funds had more of a barbell shaped risk profile with roughly 1/3 of the portfolios in lower risk agency bonds and over 1/3 of the portfolio in much higher risk ERS bonds after 2007. Given the UBS PR Funds fell by more than twice the decline in the S&P Puerto Rican Index in 2013 it appears our estimate of the true risk of the UBS Puerto Rican Bond Funds is fairly accurate.

We can see in Figure 7 that on a pre-tax basis, the UBS Puerto Rican Funds fall well below the "efficient frontier." If Puerto Rican municipal bonds received the same tax treatment as the stocks and taxable bonds plotted in Figure 1, no rational or informed investors should buy the UBS Funds because the investors could achieve the same returns with lower risk or higher returns at lower risk.

We estimate that, on a pre-tax basis, the UBS Puerto Rican Bonds Funds have the risk of a portfolio invested 85% in stock and 15% in bonds.

After-Tax Efficient Frontier

Puerto Rican residents did not pay income taxes on Puerto Rican municipal bond income but would have paid taxes on interest and capital gains paid on the stock and bond portfolios plotted in Figure 7. We adjust for this difference in tax treatment by reducing interest distributions by 33% and capital gains by 10% paid by the stock and bond portfolios. Figure 8 plots the after-tax risk and returns.



Figure 8 shows that even after adjusting for taxes, the UBS Puerto Rican Funds fall below the efficient frontier. It also shows that on an after-tax basis, the UBS Puerto Rican Funds were about as risky as a portfolio invested 95% in stock and 5% in bonds. This last point is especially dramatic given that UBS described these funds as suitable for investors interested in income consistent with capital preservation. Portfolios with the risk and return of 95% in stock and 5% in bonds are clearly not consistent with the investment objective of "income consistent with capital preservation."

I. Summary

Investors lost nearly $3 billion in 2013 in closed-end municipal bond funds managed by UBS Asset Managers of Puerto Rico and sold by UBS Puerto Rico brokers. Many of the UBS Funds lost over 50%. Over the same period, the average US municipal bond fund fell by 2.5%.

The UBS funds were extraordinarily concentrated. Securities of a single issuer accounted for over 25% - and of two issuers over 40% - of the fund's gross assets for most of the Funds. The use of leverage in the UBS funds magnified price fluctuations in the portfolio securities. Together, the leverage and concentration meant that declines in the securities of just one or two issuers had catastrophic effects on investors.

The risks UBS magnified with concentration and leverage were not just investment market risks. UBS retail brokers recommended to retail investors funds which UBS managed and that bought almost exclusively bonds UBS underwrote from issuers UBS advised. This removed important market discipline and heightened the likelihood of self-dealing.

UBS earned substantial underwriting fees from the two issuers in which it concentrated the mutual funds' portfolios. Concentrating the fund portfolios increased UBS's capacity to underwrite bonds while simultaneously minimizing sales effort. UBS also benefited from its use of leverage since it received a significant portion of the interest costs of leverage and charged fund management fees on total assets of the funds, including assets financed with leverage.