Women entrepreneurs could benefit from better targeting, without changing the program content

Every year, governments and development organizations spend billions on business upgrading programs — training, mentorship, consulting — designed to help small firms grow. These programs work, on average. For men. For women entrepreneurs, the average return is statistically indistinguishable from zero. The standard policy response has been to spend more: add gender-specific content, design women-only programs, set enrollment quotas. A new World Bank working paper offers an alternative perspective, one that centers around the cost-effectiveness. Business upgrading programs are not necessarily broken for women. The selection and targeting process could be tailored to improve their efficacy. The Gap — And What Causes It Using pooled experimental data from 2,595 entrepreneurs across Uganda, South Africa, and Mexico, the paper documents a stark divide. Without any targeting, male entrepreneurs who receive business upgrading support show large, statistically significant gains in firm performance. Women show gains indistinguishable from zero. For the average male participant, a typical program translated into measurable profit growth within months. For the average female participant, it produced no detectable change at all — a poor return on an investment that costs programs around US$758 per person. The paper's contribution is to show why this happens — and to demonstrate it can be fixed through an alternative less-expensive approach. The paper builds a Program Readiness Scorecard (PRS): a 100-point composite of ten observable behaviors and practices — the "10 Es" — grounded in the idea that returns to business upgrading are dampened when entrepreneurs face tight credit constraints or low baseline capabilities. The PRS covers three dimensions: credit constraints (Endowments, Establishment); baseline business skills (Education, Experience, Effectiveness, Exposure); and growth attitudes (Experimentation, External orientation, Evaluation). Crucially, it does not use current profits or firm size, avoiding the trap of screening already-successful firms. When the PRS is applied as a gender-neutral screen — removing the lowest-scoring entrepreneurs regardless of gender — women's returns become statistically significant. That is promising. But men's returns remain roughly twice as large. The gap narrows; it does not close. Figure 1. Rope ladder plot of intent-to-treat (ITT) estimates from pooled experimental data across Uganda, South Africa, and Mexico (N=2,595). Points show treatment effects on a standardized composite index of firm performance; vertical lines show 95% confidence intervals. The left panel shows the eligible population without targeting; the right panel shows results after gender-neutral PRS screening. The reason is structural. Men average 55.2 on the PRS; women average 51.4 — not because of innate differences in ability, but because structural barriers have limited women's access to capital, managerial experience, and professional networks. A gender-neutral cutoff, applied to an unequal distribution, still administers the program to a disproportionate share of women with lower-readiness scores relative to men. When Program-ready Women are Targeted, the Gap Closes Entirely When a higher PRS threshold is applied specifically for women — targeting those in the top quartile of the female distribution, with a score of at least 61 out of 100 — the results are striking. These are women who signal strong growth attitudes and sufficient baseline capability, even if their firms are currently small. Women above this threshold PRS achieve treatment effects virtually identical to those of the average male entrepreneur. The difference between them is negligible and statistically indistinguishable from zero. Women below this threshold continue to show returns not different from zero. The gender program gap is not an inherent feature of how women respond to business upgrading. It is a product of who gets selected into programs. Figure 2. Rope ladder plot of ITT estimates under gender-based targeting. Points show treatment effects on a standardized composite index of firm performance; vertical lines show 95% confidence intervals. High-PRS Women are those in the top quartile of the female PRS distribution (score ≥ 61). The difference in returns between Average Men and High-PRS Women is 0.013 SD (p=0.892). Smarter Targeting, Not More Spending Three policy implications follow. First, program-ready entrepreneurs are a commercially viable segment that private providers are under-serving. High-readiness women, for example, recoup the cost of a typical program within roughly four months through profit gains — the same timeframe as men. Private business development service firms, and consulting companies should treat top-quartile women entrepreneurs as an attractive market opportunity. Second, for Development Finance Institutions and grant programs, the PRS is a practical investment screen. It identifies growth potential without using firm size or revenue as proxies, which tend to reproduce existing inequalities. Applying it to pipeline selection — for value chain integration, investment readiness programs, or SME lending — would direct capital toward program-ready firms with absorption capacity, including women-led firms that are currently being passed over. Third, low-readiness entrepreneurs are not a lost cause — they need a different kind of support first. The PRS is diagnostic: it identifies the specific bottleneck holding each entrepreneur back, whether limited capital access or weak experimentation practices, pointing to targeted remedial investments that build readiness over time. The policy implication is a tiered approach to building program ready entrepreneurs: broad foundational support for many, intensive upgrading for those who are ready, with the PRS serving as the gateway between the two. The broader lesson cuts against the instinct to simply do more. Equal outcomes require acknowledging unequal starting points — and designing upgrading programs accordingly. A transparent, field-deployable scorecard can redirect billions in annual program spending toward the entrepreneurs most likely to benefit.