This is our 9th blog post for our Job Market Paper Series blog for 2025-2026.
Jiayue Zhang is a PhD student at Brown University. Her research interests are at the intersection of development, environment, and firms.
Across sub-Saharan Africa, limited access to credit is one of the major reasons why small firms struggle to grow. While recent technological innovations like pay-as-you-go (PAYG) have made it possible to finance assets that were once uncollateralizable, adoption by firms remains limited. Why don’t small firms in low-income countries adopt green technology offered on credit? An important but often overlooked aspect is that household constraints can spill into firm decisions, especially under imperfect credit market.
My job market paper strives to understand firm investment decisions given imperfect credit market, in the context of solar-credit bundles in Nakuru, Kenya. We conducted a discrete choice experiment with more than 700 small business owners by collecting their decisions on purchasing a portable solar light with a set of dynamically updated credit contracts. We then offered them the chance to purchase the solar light on credit, with random subsidies applied to either (a) down payment only or (b) flexibly on down payment or repayment. We further evaluate the impact of solar adoption on small firm performances with a six-month follow-up. The results reveal a simple but powerful insight: firm investments can be hindered by household constraints.
What We Did: Demand Elicitation with Experiment
We first collected firm owners’ discrete choices over a set of menus with different contracts. The contract down payment and repayment amounts are dynamically updated according to previous choices and a Bayesian rule, in order to effectively trace out the individual demand curve. This choice experiment is incentivized by randomly implementing their choice in one of the menus.
We then embed in this implemented menu a randomized controlled trial with two subsidy treated arms and one control arm. The implemented menus in the treated arms each include a random subsidy, either (a) applied to the down payment, or (b) flexibly applied to the down payment or repayment. This experimental design helps us to evaluate the demand elasticity with regard to the total payment, the down payment, and the future repayments.
Moving Beyond a Single Price Tag: A “Demand Surface” View
Conventionally, demand is estimated on a single price dimension (Figure 1a). But when bundling credit financing with purchases, there are multiple payments over time. Thus, total payment alone doesn’t tell the full story. We collected rich data on firm owner decisions over different price dimensions. This enables us to visualize what we call a “demand surface”, a two-dimensional contour map showing how firms’ willingness to adopt changes with both the down payment and repayment amounts (Figure 1b).
The demand surface revealed a crucial insight: at similar total prices, firms prefer payment schedules that were less lumpy or more effective in intertemporal smoothing, meaning the upfront payment and the repayments were closer in size. With a sub-sample analysis, we find that this preference is driven by firm owners who are the primary earners of their households, i.e., the household heads. Non-head owners, in comparison, behave like our textbook risk-neutral profit-maximizing firms that can effectively trade off current and future payments.
Business owners who are the primary earners of their households face a double burden: running a business while absorbing family liquidity shocks. Their investment decisions reflect the need to balance family responsibilities with business growth. In contrast, non-head owners who are often secondary earners within their households, behave more like conventional entrepreneurs. They are less constrained by household liquidity and more willing and able to trade off current and future payments. This is despite the fact that, on average, non-head owners are from households with lower per-capita income and per-capita liquidity.
This distinction blurs the traditional line between “household” and “firm.” In much of the informal economy, household liquidity constraints can spill over to business investment decisions. Policies that ignore this overlap may miss the real barriers that small businesses face when making investment decisions.
We use the treatment assignments and subsidy amount as instruments for the contract prices, to estimate price elasticity of demand. The results show that firms owned by household heads are far more sensitive to the down payment than to later repayments, whereas non-head firm owners are equally sensitive to changes in these two price dimensions. This asymmetry echoes the importance of household liquidity constraints in shaping firm investment decisions.
Treatment Effects on Adoption and Downstream Outcomes
More than 80 percent of firms offered the flexible subsidy still chose to apply it to the down payment. Even when given the option to lower their weekly repayments, most preferred to lower the upfront cost. Only 4 percent of firms in the control group purchased the solar light. A subsidy that cut the down payment increased adoption by about 12 percentage points; the flexible subsidy raised it by 9 percentage points. While we cannot reject the null that the two subsidy schemes increased adoption differently, we do find the flexible subsidy positively selecting less constrained owners into adoption. In contrast, firms whose owners are household heads, face low household liquidity and exhibit higher risk aversion are significantly less likely to adopt under the flexible subsidy scheme. This suggests that household liquidity shocks and constraints play a central role in shaping contract choice and adoption behavior in firm decisions.
As a result, firms under the flexible subsidy exhibited more pronounced post-adoption operational adjustments. They reported operating more days per month and spending more on solar energy, suggesting deeper integration of solar into their daily operations. These businesses appeared to substitute solar for unreliable grid electricity, improving their resilience to energy supply interruptions.
Both subsidy designs improved the perceived usefulness and popularity of solar. Business owners became more convinced that solar was useful and increasingly believed that other firms in their community were adopting it, reinforcing positive peer expectations and further encouraging uptake. These shifts in beliefs suggest that well-designed financing not only enables adoption but also reshapes how firms envision solar within their longer-term production strategies.
Lighting the Path Ahead
The green transition in developing economies will depend not only on technological availability but also on financial tools that make clean energy accessible. Our research shows that how subsidies are structured in contract design can matter as much as the size of the subsidy itself. As policymakers seek to scale renewable energy in contexts with credit constraints and volatile infrastructure, attention to contract design offers a practical, high-impact lever for driving sustainable and resilient growth.
Featured image: Photo taken by the author in the field, in Nakuru City of Kenya
