More Firms, Deeper Poverty? Rethinking the SME Growth Model

Governments often treat a rising number of small businesses as proof that an economy is becoming more dynamic, inclusive and resilient. But what if many of those firms are not engines of upward mobility, but signs that people have run out of better employment options?

A study titled "Entrepreneurship, Inequality, and Multidimensional Poverty: Rethinking the Role of SMEs in a Transition Economy," published in the journal Economies, examines how small and medium-sized enterprises relate to poverty and inequality across 15 regions of Kazakhstan between 2005 and 2024.

The findings challenge one of development policy's most persistent assumptions: expanding the SME sector does not automatically reduce poverty. The social value of entrepreneurship depends less on how many firms exist than on whether those firms raise productivity, create stable employment, generate meaningful value and connect workers to stronger markets.

Across developing and transition economies, SME policy is often judged by the number of businesses registered, loans disbursed or jobs nominally created. The study suggests that those metrics can conceal a deeper problem: enterprise growth may absorb unemployment without materially improving wages, reducing severe deprivation or narrowing inequality.

The SME Numbers Game Is Hiding the Real Development Story

The study separates two ideas that policymakers frequently combine: the quantity of SMEs and the quality of their economic contribution. Firm counts, SME employment and output expansion measure how large the sector has become. Gross value added, productivity and integration into wider production networks reveal whether that growth is generating stronger economic returns.

The difference is crucial. A new business may be a technology-enabled supplier entering a national value chain, or a one-person survival enterprise selling low-margin goods in a saturated local market. Both appear in official statistics as entrepreneurial activity, but their capacity to raise incomes and reduce poverty is radically different.

Using a balanced panel of regional data, the researchers assessed poverty through three measures: the proportion of people below the subsistence minimum, the average distance separating poor households from that threshold, and the severity of deprivation among the poorest. They also examined inequality through the Gini coefficient and income fund ratio.

The results show that SME indicators do not move consistently in the same direction. A higher share of SME gross value added was associated with lower poverty incidence. In one model, a 1% increase in SME value added relative to regional output corresponded to an estimated 0.61% reduction in the population living below the subsistence minimum.

The relationship supports the argument that productive firms are more likely to generate durable social gains. SMEs that produce higher value added can invest, raise wages, adopt technology, improve competitiveness and participate in supply chains beyond their immediate locality. On the other hand, a higher SME output share was associated in another model with increased poverty incidence. A 1% rise in SME output share was linked to an estimated 0.34% increase in poverty.

The study warns against treating firm creation as a universal proxy for progress. When enterprise expansion is concentrated in low-productivity, informal or low-wage activities, it may provide households with subsistence income while doing little to create lasting routes out of poverty.

Unemployment Matters More Than Business Registrations

Across the study's models, unemployment was strongly associated with poverty incidence, poverty depth, poverty severity and income inequality. A 1% increase in unemployment corresponded to an estimated 0.59% to 0.64% increase in the proportion of people living below the subsistence minimum.

The relationship remained significant when the researchers examined more intense forms of deprivation. Higher unemployment was associated not only with more people being poor, but with poor households falling further below the poverty line. The finding shifts the policy debate away from the number of entrepreneurs and toward the quality of labour-market participation. A country can create thousands of microenterprises while failing to create enough productive, formal and adequately paid work. In such a system, self-employment may function as a private coping mechanism for public labour-market failure.

The distinction is especially relevant in economies where people establish businesses out of necessity rather than opportunity. A necessity entrepreneur may earn enough to survive but lack access to finance, technology, skilled labour, larger markets or social protection. Such firms often remain small not by choice, but because their economic environment gives them few pathways to grow.

The results suggest that entrepreneurship programmes cannot substitute for employment policy. Governments still need vocational training, labour-market matching, industrial investment, regional infrastructure and support for firms capable of creating formal jobs.

The study also finds that income inequality consistently worsens poverty. Higher inequality was associated with a greater share of the population living below the subsistence minimum and with deeper deprivation among those already poor. This reinforces a broader development lesson: poverty reduction depends not only on how much income an economy produces, but on how economic gains are distributed.

Growth Can Reduce Deprivation While Widening Inequality

The study's findings on income reveal another important tension. Higher income was strongly associated with reductions in poverty depth and severity. A 1% increase in the study's income measure corresponded to an estimated 1.46% decline in poverty depth and a 0.99% decline in poverty severity. This means income growth can substantially improve the condition of households already below the poverty line. Even when they do not immediately escape poverty, they may become less deprived and more resilient.

However, the same study found that rising income was also associated with higher inequality. This is a critical warning against relying on aggregate growth as a stand-alone development strategy. Economic expansion may reduce severe deprivation while concentrating a disproportionate share of the gains among higher-income households, more productive firms or better-connected regions.

Kazakhstan's regional economy helps explain why these effects may diverge. Large mining, energy, metallurgical and manufacturing enterprises are concentrated in particular areas, reflecting both historic industrial specialisation and the uneven legacy of economic transition.

SMEs operating near these anchor industries may benefit from supplier contracts, knowledge spillovers, infrastructure and access to wider markets. Firms in regions dominated by small local services may have fewer opportunities to upgrade productivity or scale beyond subsistence activity.

The same national SME programme can therefore produce very different results depending on regional industrial structure. Many governments deploy uniform credit schemes, tax exemptions or business-registration reforms across territories with sharply different market access, skills, infrastructure and institutional capacity. Such policies may expand enterprise numbers without reducing regional inequality. Better-connected entrepreneurs can capture finance and procurement opportunities, while firms in lagging areas remain confined to local, low-value markets.

The study also reports positive associations between social assistance indicators and poverty. The authors appropriately caution that this likely reflects targeting: poorer regions receive more support. It should not be interpreted as evidence that assistance causes poverty. Still, the result points to an important distinction between reactive and preventive policy. Social protection can reduce hardship, but it cannot by itself generate productive employment, stronger wages or sustainable business growth.

Effective poverty policy thus requires both protection and transformation: support for households facing immediate deprivation, combined with labour-market, education and industrial policies that expand access to higher-value economic activity.

Stop Counting Firms and Start Measuring What They Create

Governments should stop treating SME expansion as an end in itself. Business registrations, credit approvals and enterprise counts are politically convenient because they are visible and easy to measure. But they say little about whether supported firms survive, raise productivity, improve wages or reduce poverty.

A more credible SME policy would evaluate value added, firm survival, employee earnings, formalisation, innovation, export activity and integration into larger value chains. Support should also be differentiated. A subsistence microenterprise requires a different policy package from a growth-oriented manufacturer or digital services company. Treating both as interchangeable "SMEs" produces poorly targeted programmes and misleading assessments.

Governments could prioritise technology adoption, management capability, vocational skills and supplier-development schemes that connect smaller firms with larger anchor companies. Financial support should be tied not only to starting businesses but to investment in productivity, equipment, workforce development and market expansion.

Regional policy is equally important. Areas with weak industrial bases may need infrastructure, digital connectivity and local demand creation before firm-level support can produce meaningful results. Regions with strong anchor industries may benefit more from procurement reforms, quality certification and programmes that help local SMEs enter established supply chains.

Development banks and international agencies should reconsider how they judge entrepreneurship programmes. Counting beneficiaries or loans may demonstrate activity, but it does not establish development impact. Evaluations should track whether support creates lasting employment, raises household income and benefits disadvantaged groups.

The study itself has important limitations including that it relies on aggregate regional data and cannot distinguish directly among microenterprises, innovative firms, high-growth businesses and informal operators. Nor can it separate opportunity-driven from necessity-driven entrepreneurship.

Its statistical approach accounts for common regional shocks, heteroskedasticity and serial correlation, but it does not eliminate endogeneity. Poverty may encourage business formation, just as SME development may affect poverty. The results should therefore be interpreted as conditional associations rather than proof of causality.

Some findings also require deeper investigation. SME value added was linked to lower poverty incidence but to greater poverty depth and severity in certain specifications. One possible interpretation is that productive SMEs help households near the poverty line more than those experiencing extreme deprivation. Another is that regional structures or omitted variables influence both SME performance and poverty intensity.

Future research should use household- and firm-level data, distinguish formal from informal businesses and examine gender, sector, location and entrepreneurial motivation. Comparative studies across Central Asia and other transition economies could test whether Kazakhstan reflects a wider pattern.