By Daniel Rozas,
Microfinance Focus, December 10, 2012: Over the past 18 months, one of the microfinance sector’s largest and most prominent funds, Blue Orchard’s Dexia Micro-Credit Fund (recently renamed Blue Orchard Microfinance Fund), saw a major outflow of investor capital, with some $268 million or nearly 50% of the fund’s peak value having been redeemed.  The scale of these outflows is unprecedented in the sector.  For years, investment capital largely flowed one way:  in.  The exit doors were there, but rarely used.  That is no longer the case.  The pioneer of the microfinance investment industry has now crossed another milestone in the industry’s development.

Like Dexia, many microfinance funds (commonly referred to as Microfinance Investment Vehicles or MIVs) are subject to unscheduled redemptions.  For those funds, their investors, as well as others in the sector, BlueOrchard’s experience holds important lessons, and it is those lessons that this article hopes to convey.

What caused the redemptions at Dexia?  There is no simple answer, no smoking gun.  Instead, like ocean waves that intersect at just the right time to form a rogue wave, BlueOrchard’s flagship fund appears to have been overcome by multiple factors that came together at just the wrong time.  Even so, BlueOrchard was no innocent victim.

In the beginning…

The seeds of the fund’s problems were laid at its founding.  More than just about any other microfinance investment manager, BlueOrchard saw itself as a trailblazer with a mission to “bridge the gap between microfinance and private capital.” That was in part the genesis behind its securitization deals in 2004-07. And while other MIVs were content to source large amounts of investment from development finance institutions, the Dexia fund was exclusively oriented towards private, mostly commercial money.  Indeed, BlueOrchard was successful in carrying out its mission:  the two largest Dutch pension funds – ABP and PGGM – both chose Dexia as their gateway into microfinance.

The strategy was clearly working, making the Dexia fund the 3rd largest MIV in the sector, and BlueOrchard the second largest microfinance fund manager¹.   But underlying this success was a less-than-solid foundation:  it was heavily concentrated on a relatively small number of similar investors.  When PGGM invested $41 million in March 2008, it represented some 12% of the fund’s total assets.  That same year, ABP’s investment in the fund stood at $40 million.  Combined, the two Dutch pension funds accounted for over 20% of Dexia’s total assets at year-end 2008.  While such concentration is not an issue for closed-end funds or those with long lockout periods (when investors are prevented from redeeming their shares), for an open-ended fund like Dexia, such concentration in large, similar investors posed a substantial risk.

As with many other funds, the financial crisis and subsequent downturn in the microfinance sector had its impact on the Dexia fund, which was exposed to struggling MFIs in Nicaragua, Bosnia, and Andhra Pradesh.  Even as Dexia largely met its stated return target of 1-2% over LIBOR, since 2007 Dexia has been consistently trailing most fixed-income MIVs (Figure 1).

In the earlier years of the financial crisis, Dexia actually looked very strong.  Investors valued the presumed decorrelation of microfinance from other assets, assuming that it would continue into the future.  But the story began to break down in 2010 and especially 2011 with the collapse of MFIs in Andhra Pradesh to which Dexia had significant exposure.  Indeed, it was the 1.85% write-down on Andhra-based MFIs that pulled the fund’s 2011 returns so far below its target and the sector average.

Was the fund excessively exposed to Andhra MFIs?  Not really.  At the start of the crisis in Oct 2010, 4.7% of Dexia’s portfolio was invested in MFIs operating in this Indian state.  However, some perspective is needed here:  with its 84 million inhabitants and an outstanding loan portfolio of some $1 billion, Andhra was one of the largest microfinance markets in the world, attracting investments from many microfinance funds, not just Dexia.  And it’s worth noting that the impact from what was essentially a complete collapse of this part of the portfolio was still limited – it may have largely wiped out returns for the year, but it did not reduce investors’ principal.  That’s a performance that investors holding mortgage assets or bonds of several EU countries would have been happy to replicate.

Nevertheless, Dexia’s returns should be seen in the context of increasing pressures on European institutional investors. Insurance companies and banks were buffeted by higher risk capital requirements, with microfinance often slotted in the riskiest categories. Meanwhile pension funds were facing more rigorous stress factors for unrated or illiquid assets – with microfinance likewise often qualified under both.  Most pressing of all were the low yields available in financial markets generally, and within pension funds driven by fixed return targets, low-yielding investments such as Dexia were meriting a closer look.

In the midst of all this, BlueOrchard was experiencing significant churn in its executive function.  Jean-Pierre Klumpp, CEO since 2008, left in May 2011. While his departure was unrelated to the redemptions that started a month earlier, it was also not a random event.  During the preceding months, the BlueOrchard board sought realign its two sister companies, BlueOrchard Finance (manager of Dexia and other fixed-income funds) and BlueOrchard Investments (private equity).  Subsequent organizational changes suggest that the plan didn’t succeed, at least not as initially intended: Mr. Klumpp left and Mr. Jean-Philippe de Schrevel (BlueOrchard’s founder and head of BlueOrchard Investments) took over the leadership of both companies, though his role lasted only until the end of the year.  Eventually, the two companies split for good. BlueOrchard then named one of its Board members, Marc Beaujean, as caretaker manager, before finally announcing the appointment of a permanent CEO, Wolfgang Landl, in July 2012 – 15 months after the redemptions began. During a time when its largest fund was experiencing its greatest challenges, BlueOrchard had three separate executives occupy the top post.

Riding the wave

The first flow of redemptions began in April 2011. Each month, an average of $13.5 million flowed out of the fund.  In the five months May-Sep 2011, $135 million was redeemed, constituting 24.5% of Dexia’s assets at the time.  For a fund invested in illiquid securities – loans to MFIs – handling such a stream of redemptions is an enormous challenge.  But for all its prior errors, it’s here that the fund’s investment policies paid off. Its minimum threshold for liquid assets was 10%, though in March 2011, the figure stood higher still, at 15%. This liquidity was supplemented by a standby line of credit that the fund had kept on tap for years and now was able to draw upon to help meet redemption requests. However, the most critical support came from the fund’s investment policy that set maximum loan maturity at three years and instructed the fund’s managers to ensure that the portfolio would be maturing on a rolling basis.  In practice, the policy meant that at any one time, somewhere around 25% of the portfolio would mature within the next six months (Figure 2).

When the redemptions began, the fund reacted swiftly.  In brought its disbursements to a near complete halt (Figure 3).

In April it made one loan – about $1 million – to an MFI in Zambia, thus enabling it to announce in its monthly investor update the impressive news that it was adding another country and another MFI to its portfolio.  There was no mention of the $7 million that had already been redeemed, and another $30 million that was about to be redeemed (due to the 30-day notice requirement, BlueOrchard knew redemption levels a month in advance).  During the subsequent months, BlueOrchard continued to make a handful of disbursements, which helped it maintain some level of activity and provided material for its investor reports. But this activity was but a shadow of its former self.  During the 19 months from April 2011 to October 2012, the fund disbursed a total of $30.5 million – less than what it had disbursed during the single month of March 2011. But by not disbursing, BlueOrchard was able to redirect nearly the entire reflow stream from the fund’s maturing loans to meet investor redemption demands.

BlueOrchard had one other option at its disposal: if any one month’s redemptions in the fund exceeded 10% of shares outstanding, its Board of Directors could elect to defer a redemption request.  It’s a testament to how well the fund was able to manage the outflows that this option was never exercised.

Learning the right lessons

The most difficult case studies are those where events hinge on multiple factors.  One can certainly describe the factors, but that doesn’t mean all are equally relevant to answering the key questions: what should the institution have done differently?  And what should others take away from the case to avoid a similar fate?

Credit risk management. Let’s take the most obvious and well-known factor:  the Andhra Pradesh crisis.  Dexia’s losses from this were large, wiping out nearly its entire returns for 2011. But does that make it relevant?  Should BlueOrchard have stayed away from that market?  In hindsight, certainly.  Even looking ahead, one could’ve seen trouble.  Still, investing is risky.  Being hit by unexpected losses – even large ones – is an outcome one can never entirely eliminate.  And the fund’s position in Andhra Pradesh was not unreasonable, given both the size and widely-assumed potential of that market.  By no means should one underestimate the importance of the Andhra crisis to the microfinance sector generally and to microfinance investors specifically, but in this particular case, for all the losses and negative publicity that it generated, Andhra was not ultimately the deciding factor behind the redemptions at Dexia.

Low return target. A more relevant factor than exposure to Andhra was the fund’s low return target, which became especially acute in a low-rate environment.  When rates were high, Dexia was in line with or outperforming many of its competitors, but once LIBOR started coming down, the fund’s strategy all but guaranteed that its returns would trail those of other MIVs.  Already in 2010-11, the long-term outlook for interest rates was that they would remain low for at least several years, so it’s worth questioning whether the fund should have revisited its return target at the time.  Instead, after a decade of solid returns, BlueOrchard appears to have grown complacent on this point.  At a minimum, it should have recognized that maintaining a low return target would expose it to greater volatility, as commercial investors cashed out to seek higher yields.  Then again, benchmarking to a variable return target is in the may not be an optimal strategy for a fund whose portfolio consists of illiquid, mostly fixed-rate loans to MFIs, especially when that fund’s investors are commercial institutions that have their own fixed return targets to meet.

Excessive investor concentration. In light of this return target and the open-ended structure of the fund, the fact that Dexia had a concentrated funding base was a high risk.  Because some of these investors shared strong similarities in form (pension funds) and geography (Netherlands), the risk of large-scale redemptions was further increased. Indeed, when the first redemptions began, they may well have sparked a run of sorts – not a run because investors were concerned about not being able to redeem, but rather, a run based on basic herd behavior common in financial markets.  When facing a credit committee, a portfolio manager may find it easier to close a position than to defend it, especially when it’s generating unwanted attention due to its low yield (both in relative and absolute terms), negative headlines in the press, and a name that happens to be shared with a bank that just collapsed (Dexia).  This is all the more true when that position has been held for some 4-5 years and has already yielded good returns in the past.  In such a context, the argument for closing out the position is likely to outweigh the argument for staying put, especially if other pension funds in the same country have already exited.

Poor governance. Finally, there is the question of management churn.  This, frankly, is a self-inflicted wound – and a grievous one at that.  I’m not questioning the realignment itself; it may well have been the right strategy in the long-run.  However, the effort was terribly ill-timed. Recall that the Andhra Pradesh crisis exploded in October 2010.  The Board surely was aware of Dexia’s portfolio exposure to the region, and was presumably also aware of the media controversy that was being played out as a result. This was a time that called for stability in the management suite. Instead, to this combustible mix the Board’s efforts added an unnecessary and, ultimately, deeply damaging aura of organizational instability and uncertainty.

In fact, there was a strong case to be made to investors to remain with the fund.  By mid-2011 – still early in the redemption cycle – the fund’s difficulties were largely behind it: the Andhra losses were already provisioned for, and the fund made solid annualized returns of 1.86% in Aug-Dec 2011 and 3.02% in Q1 2012.  By building on this and laying out a meaningful, long-term vision for the fund, a capable, committed manager might well have been able to stem the tide of redemptions. Unfortunately, no matter how convincing the message, it cannot be delivered credibly by a caretaker CEO.

Governance has for some time been emphasized as an important component of MFI risk management. Presumably, the same arguments should also apply to MIV managers. In BlueOrchard’s case, governance quality fell short. One imagines that a BlueOrchard credit committee would not have been very tolerant of such management churn within an investee MFI, especially if it were happening in a time of crisis, and rightly so. The organization would benefit from having those same standards applied to its own Board.
***
¹. Xavier Reille, Forster, Rozas:  Foreign Capital Investment in Microfinance: Reassessing Financial and Social Returns, CGAP, May 2011.

Note to readers:  due to the sensitive nature of the topic, unless otherwise noted, the information contained here was received from anonymous sources familiar with the situation. To the extent possible, it was cross-referenced with publicly-available information, including that provided by BlueOrchard upon author’s request. To all who contributed information, you have my deep appreciation. Thank you.

In addition, BlueOrchard requested that the following statement be included in the article: “BlueOrchard has declined to comment on alleged investors in the fund as it is bound by professional secrecy and confidentiality obligations. Furthermore, BlueOrchard has declined to comment on any view expressed in this article by its author.”