Microfinance is among the oldest strategies in impact investing and one of the few with a record long enough to be judged on financial terms alone. Tameo Impact Fund Solutions has calculated its Microfinance Private Debt Index (TX-MFD) from the monthly net asset values of private debt funds lending to microfinance institutions in emerging and frontier markets since 2004 (Tameo Impact Fund Solutions, 2025). A study published by the St. Gallen consultancy c-alm in August 2026 uses this record to ask whether microfinance belongs in the fixed income allocation of a regulated Swiss investor (Beiner & Weder, 2026). The study deliberately sets aside the question of social impact, on which the academic evidence remains heterogeneous, and concentrates on return, credit risk, crisis behaviour and diversification.
Swiss investors typically gain exposure through a microfinance investment vehicle (MIV). Around 90% of a fund's assets are lent to microfinance institutions (MFIs), often some 150 of them, with the remainder held in money market instruments or bonds. Loans run for two to three years, amortise on a fixed schedule and are predominantly senior; roughly 60% carry a fixed rate and a large share is denominated in US dollars (Beiner & Weder, 2026). Each MFI then divides its borrowing into many small loans to households and micro-entrepreneurs. Vehicles differ considerably in their currency and interest rate hedging, in the mix of senior and subordinated debt and in the size of the institutions they finance – differences that determine the risk profile an investor actually acquires.
At the end of the chain, the economics are instructive. Informal moneylenders charge low-income borrowers between 40% and 200% a year, whereas MFIs typically charge 25% to 30% (Banerjee & Duflo, 2011, as cited in Beiner & Weder, 2026). The difference rests on lending techniques refined over decades: standardised products, frequent repayments, peer accountability and dynamic incentives, under which access to larger loans depends on the timely repayment of smaller ones. Randomised trials summarised by Cai et al. (2025, as cited in Beiner & Weder, 2026) find such incentives particularly effective in reducing default among higher-risk borrowers. The market remains small: c-alm estimates assets in MIVs at USD 20–25 billion, against some USD 1.57 trillion in impact investing overall (Global Impact Investing Network, 2024, as cited in Beiner & Weder, 2026).
The evidence and its limits
Between the end of 2003 and the end of 2025, the CHF-hedged index returned around 1.3% a year. Swiss mortgages and hedged investment-grade foreign-currency bonds earned the same, local-currency emerging market government bonds slightly less, and Swiss franc bonds somewhat more. Asset classes with a sub-investment-grade component earned considerably more: 3.3% a year for senior secured loans, 3.6% for hard-currency emerging market bonds and 4.4% for US high yield (Beiner & Weder, 2026).
What distinguishes microfinance is its annual volatility of 0.6%, comparable to the KGAST real estate index and half that of Swiss mortgages. Its ratio of return to volatility, at 1.9, is the highest among the asset classes examined; hedged investment-grade foreign-currency bonds reach 0.4 (Beiner & Weder, 2026). The study is explicit that this figure is not a measure of economic risk. Monthly NAV pricing, ex-ante provisioning for expected losses and the patience of the sector's development lenders all smooth reported returns. For a pension fund, smoothing is nonetheless not without value: it translates into a steadier funding ratio and, over time, into more predictable crediting of members' retirement savings.
Credit quality must be inferred, since MFI loans carry no external ratings. MIVs have provisioned on average around 0.4% of assets a year for expected losses, while realised write-offs have averaged about 0.2%, with cumulative provisions consistently above write-offs. Applying an assumed recovery rate of 50% – in line with senior unsecured corporate debt – c-alm maps the implied default experience against Moody's historical default rates and places it between the BBB and BB categories (Beiner & Weder, 2026). Two mechanisms explain the low losses. MFIs absorb defaults first, through loan-loss reserves and equity that averages about 20% of assets against roughly 5% at commercial banks (Rodríguez Monroy & Huerga, 2010, as cited in Beiner & Weder, 2026). Troubled loans are, moreover, usually restructured or refinanced rather than written off, and because each MFI borrows from several funds, no single vehicle bears a loss in full.
The crisis record is the strongest part of the case. During the global financial crisis, sub-investment-grade bonds lost more than 30% and investment-grade bonds around 3%, whereas the microfinance index fell by no more than 0.5%. Its maximum drawdown in the pandemic was 1.0% (Beiner & Weder, 2026). Between February and April 2020, MFI lending fell by around 80% and repayments by about half (Lee et al., 2025, as cited in Beiner & Weder, 2026). Rather than enforce claims bilaterally, MIVs signed coordination agreements, extended maturities and rescheduled loans, while development finance institutions (DFIs) and regulators provided liquidity and moratoria (Mendelson & Rozas, 2024, as cited in Beiner & Weder, 2026). Some smaller MFIs did not survive, but losses to investors remained limited.
The limits are equally clear. The index comprised only six funds in 2025, is equally weighted and may be affected by survivorship bias. Pricing lags listed markets, as the delayed pandemic markdown in April 2020 showed. The worst month in two decades, a loss of 0.8% in 2013, arrived only when one fund wrote off loans exposed to the 2010 Andhra Pradesh crisis after the customary two to three years of restructuring – a reminder that losses in microfinance are deferred rather than absent (Beiner & Weder, 2026). Political events, fraud and armed conflict produce lower recoveries than economic defaults and, in extreme cases, total losses.
The market-conform question
For Swiss pension funds, whether an investment earns a market-conform return is a regulatory obligation under art. 71 BVG and art. 50–52 BVV 2 rather than an academic question. c-alm acknowledges that the answer depends on both the period and the benchmark. Over twenty years, microfinance matched local-currency emerging market government bonds but fell well short of hard-currency emerging market bonds. Over the past decade it trailed both, while over the six years to 2025 its short duration protected it far better than listed bonds as interest rates rose (Beiner & Weder, 2026).
The study therefore turns to risk-adjusted returns and expected credit losses, and on that basis concludes that commercial microfinance can be classified as market-conform. Its forward-looking building-block estimate, as at the end of 2023, puts the expected net return of CHF-hedged microfinance at around 1.5% after expected losses, currency hedging and running costs – comparable to hedged hard-currency emerging market bonds at 1.7% and clearly above hedged investment-grade foreign-currency bonds at 0.4% (Beiner & Weder, 2026). The cost of hedging into Swiss francs, estimated at around four percentage points, is the decisive deduction.
There is a structural reason why returns are not higher. DFIs pursue developmental as well as financial mandates and often lend to MFIs on terms commercial funds cannot match; in some markets they crowd private lenders out entirely (Beiner & Weder, 2026). Their presence stabilises the sector, as the pandemic demonstrated, but it also compresses the return available to private capital. Investors in microfinance are, in effect, buying the stability that the developmental lenders help to provide.
Translating the evidence into Swiss practice
To test the diversification argument, c-alm modelled a simplified average Swiss pension fund allocation derived from the UBS pension fund performance index, with 34% in fixed income, 40% in equities and 26% in Swiss real estate. Replacing part of the hedged investment-grade foreign-currency bonds with microfinance lowered portfolio volatility while leaving returns essentially unchanged. Moving from an equal split between the two to a full substitution reduced volatility by 15 basis points at a cost of less than one basis point of return (Beiner & Weder, 2026). The effect grows with the size of the fixed income allocation and disappears where there is none.
The benefit depends on what microfinance replaces. Substituted for assets with similar historical returns – investment-grade foreign-currency bonds or local-currency emerging market debt – it preserves the return profile. Substituted for high yield or senior secured loans, it lowers risk but also lowers return (Beiner & Weder, 2026). For the board, the relevant question is therefore not whether microfinance is attractive in isolation but which position in the existing fixed income portfolio it is meant to replace.
Implementation deserves the same attention as allocation. The index is a guide to the asset class, not to any individual vehicle, and vehicles differ in their exposure to local currencies, in whether fixed-rate loans are hedged against interest rate risk, and in their concentration on larger tier 1 institutions or smaller, faster-growing ones. Illiquidity must be assessed within the liquidity budget of the whole portfolio rather than position by position. Due diligence should accordingly cover the manager's provisioning and write-off policy, its recovery capability and its participation in creditor coordination – the mechanisms on which the asset class's record has depended.
For Swiss pension funds, microfinance is thus less a return engine than a stabilising component for fixed income. Its case rests on credit quality close to investment grade, low dependence on listed markets and two decades of composure through crises, all of which reflect the structure of the sector rather than chance. Whether that case holds for a particular fund depends on the role assigned to the allocation, the asset it replaces and the vehicle chosen to implement it.
Questions for 10 November
The Swiss Institutional Symposium will bring together asset owners for whom the composition of fixed income is a recurring question. Three questions follow from the evidence:
- Does smoothed volatility have a legitimate value for a pension fund’s funding ratio, or should it be adjusted for in risk budgeting?
- How should a board document a market-conform return under BVV 2 for an asset class without external ratings or listed prices?
- Where does the presence of development lenders end as a source of stability and begin as a constraint on commercial returns?
References
Beiner, S., & Weder, P. (2026, August 21). Mikrofinanz: Kleine Kredite, grosse Stabilität [Microfinance: Small loans, great stability] [Asset class analysis]. c-alm AG. https://cdn.website.prod.eng.c-alm.ch/PUBLICATION/Mikrofinanz~_Kleine_Kredite~_grosse_Stabilitaet_DE.pdf · Bundesgesetz über die berufliche Alters-, Hinterlassenen- und Invalidenvorsorge (BVG) vom 25. Juni 1982, SR 831.40. · Tameo Impact Fund Solutions. (2025). Microfinance Private Debt Index (TX-MFD) [Data set]. https://tameo.solutions/analytics · Verordnung über die berufliche Alters-, Hinterlassenen- und Invalidenvorsorge (BVV 2) vom 18. April 1984, SR 831.441.1. Note. Works marked «as cited in» in the text are secondary citations drawn from Beiner and Weder (2026): Banerjee and Duflo (2011), Cai et al. (2025), Global Impact Investing Network (2024), Lee et al. (2025), Mendelson and Rozas (2024) and Rodríguez Monroy and Huerga (2010). Performance data are licensed from Tameo Impact Fund Solutions.