The underground economy that sustains half the world’s poorest countries

The numbers are precise enough to be insulting. In Lima, Peru, Hernando de Soto’s research team spent nine months of near-full-time work navigating the bureaucratic requirements to register a single small garment workshop. The legal cost came to $1,200 — thirty times the monthly minimum wage. Not the cost of running the workshop. The cost of being allowed to exist in the formal economy’s ledgers.

This isn’t an anecdote about red tape. It’s arithmetic. A worker at the bottom of the income distribution, attempting to legitimize a business that might employ three people sewing garments in a Lima backstreet, faces an entry price that represents two and a half years of baseline wages — before a single thread is bought, before a single hour of labour is paid. The rational response to that price is obvious. You stay informal. You stay invisible. You build something real inside the system’s blind spot.

The question that follows isn’t about the workers who made that calculation. It’s about the system that set that price. Who designed the entry requirements at $1,200? Who was the intended customer? And what does the existence of hundreds of millions of similar calculations, made across five continents over decades, tell us about the formal economy’s actual function?

The scale

Before arguing about what the informal economy means, it’s worth establishing what it is — and demolishing the frame that treats it primarily as a site of evasion, crime, or failure.

The International Monetary Fund’s working paper by Leandro Medina and Friedrich Schneider (WP/18/17, 2018), covering 158 countries from 1991 to 2015, found an average shadow economy of 31.9% of GDP. That’s the average. In developing nations, the World Bank’s foundational 2010 dataset by Schneider, Buehn, and Montenegro put the figure at around 41% for 1999–2000. Bolivia’s informal economy measured 62.3% of GDP in the Medina/Schneider dataset. Zimbabwe’s sat at 60.6%. These aren’t countries on the margins of the global economy. These are countries where informality is the economy.

The World Bank’s 2021 report “The Long Shadow of Informality” found that informal employment accounts for more than 70% of total employment in emerging and developing economies. In Sub-Saharan Africa, the informal economy contributes roughly 36% of GDP — and that figure almost certainly understates the reality, given the measurement problems inherent in counting activity that exists precisely because it isn’t counted. The ILO’s Third Edition of “Women and Men in the Informal Economy: A Statistical Picture” (2018) found that 92% of employment in Sub-Saharan Africa (excluding Southern Africa) is informal. For women specifically, the share is higher still.

These numbers encompass street vending, subsistence farming, domestic labour, the construction crews that build the apartment blocks housing the urban middle class, and the transport networks that move people when no licensed service will. They do not encompass money laundering. They do not encompass organised crime. They do not encompass the elaborate offshore structuring by which corporations and wealthy individuals shift profits across borders to minimize tax exposure. The informal economy, as measured and as discussed here, is the economy of the poor — not a shadow cast by the criminal, but the primary light source for half the world.

That moral charge has been persistently misapplied. Informal workers are regularly framed as cheating — evading their obligations to the social compact, refusing to contribute to the systems that sustain shared public goods. The empirical record inverts this. The people selling tomatoes from a roadside table in Lusaka or driving a shared taxi in Lagos are not gaming a system they belong to. They are building a parallel system because the official one either actively excluded them or priced entry beyond their reach.

The measurement problem

Measuring an economy that by definition avoids official recording is methodologically treacherous. The dominant approaches include MIMIC models (Multiple Indicators Multiple Causes), which infer shadow economy size from observable correlates like cash holdings, electricity use, and labour market participation, and currency demand approaches, which estimate informal activity by tracking excess money supply. Both methods are defended and attacked in the literature. The Medina and Schneider IMF WP 18/17 (2018) paper uses a dynamic version of the MIMIC approach. The most substantive published critique comes from Edgar Feige — his 2016 rejoinder "Professor Schneider's Shadow Economy: What Do We Really Know?" (Journal of Tax Administration, Vol. 2, No. 2) found MIMIC estimates to suffer from conceptual flaws, arbitrary data manipulation, and insufficient documentation for replication. What all parties agree on: the informal economy is large, its measurement is imprecise, and the direction of the bias in official estimates is almost certainly toward undercounting rather than overcounting.

The logic of exclusion

De Soto’s Lima finding was a structural exhibit, not an outlier. His argument in “The Other Path” (1986) and “The Mystery of Capital” (2000) was that the high cost of legal registration wasn’t random — it reflected institutions built to serve elites and pre-existing economic actors who had an interest in limiting competition. Informality, on this reading, was not the failure of aspirational workers to join the formal economy. It was the predictable result of formal institutions that operated as exclusion mechanisms.

The academic literature since has refined and largely confirmed the causal structure. Work by Fracasso, Vittucci Marzetti, and Coletto in the Review of Economics and Institutions (Vol. 9, No. 1, 2018) found that informality is positively associated with institutional extractiveness, with trust operating as the causal mechanism: where people experience formal institutions as extraction rather than service, the rational response is to route as much economic life as possible outside those institutions. The direction of causality matters. It runs from institutional quality to informality rates — not the reverse.

The cost structure of formalization bears examination. Work by Jara, Deza Delgado, Oliva, and Torres (2023) found effective formalization tax rates ranging from 8.5% to 42% across Latin American countries, with the highest rates falling on low-earning self-employed workers — precisely the people for whom the margin between staying in business and failing is thinnest. The burden is not evenly distributed. The formalization cost, expressed as a share of earnings, is highest at the bottom of the income distribution and falls as income rises. This isn’t a neutral tax — it’s a regressive toll on economic participation.

De Soto’s “dead capital” argument extended the analysis: informal workers might hold assets — homes, workshops, equipment — but without formal title or registration, those assets couldn’t be leveraged as collateral. The formal credit system was inaccessible. The empirical record on whether formalization reliably delivered credit access after the fact is mixed; land titling programmes in Peru and elsewhere did not consistently translate into bank lending to newly titled owners. The promised benefit of formalization — integration into capital markets — largely didn’t materialize for workers at the bottom of the distribution. What materialized was a tax obligation.

The justice system’s inaccessibility runs parallel. The World Justice Project’s “Global Insights on Access to Justice 2019” — surveying more than 100,000 people across 101 countries — found that just 17% of respondents who faced a legal problem took it to any authority or third party for mediation or adjudication. Not 17% to formal courts. 17% to any third party at all. The other 83% either handled problems through direct negotiation, gave up, or absorbed the loss. For workers in informal settings — where there is no employment contract, no enforceable lease, no legal record of a transaction — this isn’t an abstract problem. It’s the material condition of their working lives every day.

What the informal economy actually builds

The sophistication of informal financial systems is consistently underestimated by people who have never needed them.

Rotating savings and credit associations — ROSCAs — appear in virtually every region of the developing world under different names. Tontines in West Africa. Chamas in Kenya. Hui in China. Tandas in Mexico. Gameya in Egypt. The mechanism is invariant: a group of people pool fixed contributions at regular intervals, and each member takes the lump sum in rotation. No interest. No collateral requirement. No credit history check. The commitment device is social — exit is possible but costly in reputational terms. Mary Kay Gugerty’s study of ROSCAs in western Kenya, published in Economic Development and Cultural Change (Vol. 55, 2007), found precisely this: the social structure of the ROSCA provides the commitment mechanism that enables people with time-inconsistent preferences to save at all.

This is not a primitive substitute for banking. For people who cannot access banks — because they lack the documentation, the minimum balance, or a branch within reasonable distance — ROSCAs are a sophisticated institution adapted to actual constraints. Gugerty’s title, “You Can’t Save Alone,” names the underlying logic with precision.

The hawala system — documented in IMF Occasional Paper No. 222, “Informal Funds Transfer Systems” — operates on a similar principle of trust-based obligation. A worker in the Gulf states with family in Pakistan or Bangladesh can transfer money through hawala brokers faster, more reliably, and at lower cost than through any formal banking channel available in the same setting. The IMF’s own analysis acknowledges the system’s functional advantages in regions where formal financial infrastructure is inadequate. The paper was written largely in the context of post-9/11 regulatory concern, but its documentation of hawala’s operational efficiency is unambiguous.

Informal food systems sustain urban populations in sub-Saharan Africa at a scale that no formal retail chain approaches. The GAIN Health report “Informal food retail in urban areas” (2020) documented that in cities including Maputo, Harare, Lusaka, and Blantyre, over 90% of households visit informal retail points at least weekly. The formal supermarket is a middle-class institution, geographically and economically out of reach for the majority of urban residents. The informal food vendor outside the bus station, the woman selling cooked food from a table near the market — these aren’t gaps in the food system. They are the food system.

Informal construction crews build the cities of the global South. The apartment buildings, the market structures, the incremental self-built housing that houses the urban majority in cities from Accra to Nairobi to Dhaka — these are built by unlicensed workers with skills developed outside any formal credential system. The buildings often work. The workers remain invisible.

None of this should be read as romanticising the informal economy. The costs are real and they fall hardest on the people with the least capacity to absorb them. There is no enforceable contract when an employer decides not to pay. There is no sick pay when a street vendor gets ill and can’t work. There is no pension accumulating for thirty years of labour. There is no legal recourse when a landlord seizes goods. Women are overrepresented in the most precarious informal sectors — domestic labour, street food vending, piece-rate home production — where working conditions are least visible and legal protection is most absent. The ILO’s 2018 data shows women constitute a higher share of informal employment than men across Sub-Saharan Africa. The informal economy is a system that sustains and exploits at the same time, and it is primarily the people with the fewest alternatives who experience both simultaneously.

M-Pesa and the formal-informal boundary

In 2016, Tavneet Suri and William Jack published findings in Science (Vol. 354, pp. 1288–1292) from a long-run study of M-Pesa, Kenya's mobile money platform. Access to M-Pesa increased per capita consumption and lifted approximately 194,000 Kenyan households out of poverty. The gender effects were significant: roughly 185,000 women shifted from subsistence farming into business or retail work, enabled by the financial infrastructure M-Pesa provided. M-Pesa is neither entirely formal nor informal — it operates within a licensed regulatory framework but serves populations that formal banking had not reached. It demonstrates that the formal-informal boundary is a spectrum, and that innovations at that boundary can deliver material gains without requiring full institutional integration. The tool was adopted because it was useful, not because anyone mandated its use.

The formalization agenda — whose project is this?

The case for formalization, stated at its best, is coherent. The ILO’s Recommendation 204 (2015), “Transition from the Informal to the Formal Economy,” explicitly frames the goal as extending social protection and rights to workers — not primarily as a revenue project. The argument that formalization, where institutions are functional and trust is adequate, can deliver real improvements in worker welfare is not wrong. In contexts where courts work, where social protection systems actually pay out, where the cost of registration is proportionate to income, formality can be a genuine upgrade.

The problem is that those conditions are absent at the bottom of the distribution in exactly the countries where formalization campaigns are most actively promoted.

David McKenzie and Yaye Seynabou Sakho’s study of Bolivian firms — published in the Journal of Development Economics (Vol. 91, No. 1, 2010) — provides the clearest empirical evidence of the distributional problem. Bolivia has among the highest informality rates in Latin America. The authors used distance from a tax registration office as an instrument for formalization, identifying causal effects on firm profits. The finding: tax registration raised profits for mid-sized firms in the sample. But it lowered profits for both the smallest and the largest firms. The profitability window for formalization is narrow, concentrated in a middle band that excludes the most marginal operations — which is to say, the operations whose workers most need the social protection formalization is supposed to deliver.

This isn’t a Bolivian peculiarity. The same analysis by Jara, Deza Delgado, Oliva, and Torres on formalization tax rates shows the same structure: the effective burden falls heaviest on those with the smallest margins. The fiscal arithmetic of formalization is regressive by construction. The state captures revenue from the formalised worker. The social protection benefits that formalization is supposed to unlock arrive — when they arrive at all — on a different timeline, through different institutions, subject to different political priorities.

Vietnam makes the asymmetry legible. Research published in the Asian Development Review (Vol. 37, No. 1, 2020) by Amadou Boly found that formalization increased the likelihood of tax payment by 20% and the amount of taxes paid by 93%. Revenue gains from formalization are immediate and measurable. They show up in the next budget cycle. The social protection side of the ledger tells a different story. Across East and West Africa, contributory pension systems cover less than 10% of the labour force in the majority of countries surveyed — the World Bank’s 2019 analysis found this to be true in two-thirds of Sub-Saharan African countries. These are the countries — and these are the workers — to whom the formalization agenda promises integration into social protection systems. The pension coverage data suggests what those systems actually deliver.

The World Bank’s 2021 “Long Shadow of Informality” report added a dimension that complicates the familiar buffer-stock theory of informal employment. Conventional wisdom held that the informal sector served as a counter-cyclical absorber — when the formal economy contracted, displaced workers fell into informality, cushioning the blow. The World Bank’s finding undermines this. During recessions, the informal sector does not absorb the displaced; it contracts in parallel with the formal economy. The supposed buffer doesn’t buffer. What exists is not a stable alternative to formal employment but an economy that moves in sync with the formal cycle while offering none of its protections.

Name the state’s interest plainly: revenue from formalization is a number that goes in a budget forecast. The gains from improving court systems, reducing corruption, strengthening social protection administration — these are diffuse, long-lagged, politically expensive, and cannot be expressed as a budget line. Formalization drives raise revenue. Institutional reform doesn’t have a clean model. The asymmetry explains why states consistently choose registration campaigns over the harder work of institutional improvement. It is not a mistake. It is a rational choice made by institutions with their own interests, interests that don’t align with the workers being formalized.

ILO Recommendation 204 and the gap between commitment and delivery

The ILO's Recommendation 204, adopted in 2015, is the international community's most substantive commitment to managing the transition from informal to formal employment. Its framework explicitly prioritises rights before revenue — social protection extension, decent work standards, and worker organization before tax registration. The document is real. The commitment is formal. The problem is implementation: national programmes styled as responses to Recommendation 204 have consistently led with registration and tax compliance components, because those are legible to treasury ministries and measurable in annual budget reviews. The rights and protection components require institutional investment that doesn't produce a near-term measurable return. Recommendation 204 is a useful benchmark precisely because the gap between what it calls for and what national programmes deliver is so consistently wide. That gap is the policy problem.

The institutional failure at the centre

Reclassifying a worker — giving them a tax ID, adding them to a formal register — does not change the institutions that made informality rational. The causal arrow runs from institutional quality to informality, not the reverse. A worker who gets a registration number ends up with a tax obligation and without the courts, the credit, or the social protection that formal status nominally confers. The material position is worse, not better.

Daron Acemoglu, Simon Johnson, and James Robinson’s foundational American Economic Review paper (Vol. 91, No. 5, 2001) established the mechanism. Using settler mortality as an instrument for institutional quality — the key insight being that Europeans built extractive institutions in high-mortality colonies and settlement-oriented institutions in low-mortality ones — they demonstrated the persistent effect of colonial institutional origins on current economic outcomes. Extractive institutions were built to transfer resources from the colony to the metropole, not to provide services to the population they governed. Trust in those institutions is chronically low because the institutions historically didn’t serve the people they now govern. That history isn’t a grievance; it’s a causal explanation.

The countries that have genuinely reduced informality at scale did so through institutional improvement, not registration drives. Georgia after the Rose Revolution offers the clearest recent illustration: following the 2004 Saakashvili-era reforms, the government undertook substantive institutional restructuring — tax simplification, anti-corruption measures, administrative overhaul. Documented reductions in employment informality followed over the subsequent decade. The cause was not a registration campaign. It was a change in what formal institutions actually offered — lower costs, less extraction, marginally more trust. That sequence runs consistently through the available evidence: institutional quality first, formalization later, as a consequence rather than a cause.

The political economy of why states prefer the registration campaign over the harder work is not mysterious. Revenue is forecastable. Courts are not. A simplified business registration portal generates a revenue number that goes in next year’s budget proposal and can be presented to the IMF as fiscal improvement. Reducing corruption in the judiciary produces gains that are diffuse, long-lagged, contested in measurement, and politically dangerous because they implicate powerful actors. The preference for legible revenue over genuine reform is not a failure of understanding. It is a rational response by bureaucracies and governments to their own incentive structure.

The gap between what multilateral institutions say about informality — the World Bank’s conceptual frameworks, the ILO’s recommendations, the IMF’s technical assistance papers — and what national programmes actually deliver is not accidental. It reflects the fundamental conflict between the stated goal (improving conditions for informal workers) and the operational goal (expanding the state’s revenue base). Both are real goals. One gets funded.

Closing

Return to Lima. The $1,200 and the nine months were not a design flaw in an otherwise functional system. They were the accurate price of a system built for someone else — for the established commercial interests, the educated professionals, the already-capitalized businesses that could absorb entry costs as a rounding error. The garment workshop in the backstreet was never the intended customer.

The persistence of the informal economy across five continents and several generations of development policy is sometimes described as a puzzle. It isn’t. It is the legible outcome of formal institutions that have not offered — and in many cases were not designed to offer — a better deal to the majority of the population. The people inside the informal economy did not choose it out of backwardness or evasion. They chose it because the arithmetic made sense, and because the alternatives on offer repeatedly failed to deliver what was promised.

The standard policy prescriptions — simplified registration, reduced compliance costs, tax enforcement, mobile-first bureaucracy — reduce friction at the margin. But they do not touch the structure that makes the margin the decisive variable in the first place. A cheaper registration portal does not fix courts that don’t work. A tax ID does not create a pension system that pays out. Reduced entry costs do not produce the institutional trust that makes formality worth the price of admission.

The problem, stated precisely, is not that informal workers resist integration into the formal economy. It is that the formal economy has not yet built the institutions that would make integration worth wanting. Until that changes, the informal economy will not shrink. It will adapt.

Descargo de responsabilidad de Gen AI

Algunos contenidos de esta página han sido generados y/o editados con la ayuda de una IA Generativa.

Medios de comunicación

Himanshu Singh – Pexels

Principales fuentes y referencias

Medina, Leandro and Friedrich Schneider. “Shadow Economies Around the World: What Did We Learn Over the Last 20 Years?” IMF Working Paper WP/18/17, International Monetary Fund, 2018.

Schneider, Friedrich, Andreas Buehn, and Claudio E. Montenegro. “Shadow Economies All over the World: New Estimates for 162 Countries from 1999 to 2007.” World Bank Policy Research Working Paper No. 5356, World Bank, 2010.

World Bank. “The Long Shadow of Informality: Challenges and Policies.” World Bank Group, 2021.

International Labour Organization. “Women and Men in the Informal Economy: A Statistical Picture.” Third Edition. ILO, Geneva, 2018.

De Soto, Hernando. “The Other Path: The Invisible Revolution in the Third World.” Harper & Row, 1989. Originally published in Spanish as El Otro Sendero, Lima, 1986.

De Soto, Hernando. “The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else.” Basic Books, 2000.

Fracasso, Andrea, Giuseppe Vittucci Marzetti, and Diego Coletto. “Informal Economy and Extractive Institutions.” Review of Economics and Institutions, Vol. 9, No. 1, 2018.

Jara, H. Xavier, María Cecilia Deza Delgado, Nicolás Oliva, and Javier Torres. “Financial Disincentives to Formal Employment and Tax-Benefit Systems in Latin America.” International Tax and Public Finance, Vol. 30, No. 1, 2023.

World Justice Project. “Global Insights on Access to Justice 2019.” World Justice Project, Washington DC, 2019.

Gugerty, Mary Kay. “You Can’t Save Alone: Commitment in Rotating Savings and Credit Associations in Kenya.” Economic Development and Cultural Change, Vol. 55, No. 2, pp. 251–282, University of Chicago Press, 2007.

El Qorchi, Mohammed, Samuel Munzele Maimbo, and John F. Wilson. “Informal Funds Transfer Systems: An Analysis of the Informal Hawala System.” IMF Occasional Paper No. 222, International Monetary Fund, 2003.

GAIN (Global Alliance for Improved Nutrition). “Informal Food Retail in Urban Areas.” GAIN Health, 2020.

Boly, Amadou. “The Effects of Formalization on Small and Medium-Sized Enterprise Tax Payments: Panel Evidence from Viet Nam.” Asian Development Review, Vol. 37, No. 1, pp. 140–158, MIT Press / Asian Development Bank, 2020.

McKenzie, David and Yaye Seynabou Sakho. “Does It Pay Firms to Register for Taxes? The Impact of Formality on Firm Profitability.” Journal of Development Economics, Vol. 91, No. 1, pp. 15–24, 2010.

Suri, Tavneet and William Jack. “The Long-Run Poverty and Gender Impacts of Mobile Money.” Science, Vol. 354, No. 6317, pp. 1288–1292, 2016.

Acemoglu, Daron, Simon Johnson, and James A. Robinson. “The Colonial Origins of Comparative Development: An Empirical Investigation.” American Economic Review, Vol. 91, No. 5, pp. 1369–1401, 2001.

International Labour Organization. “Transition from the Informal to the Formal Economy.” Recommendation 204. ILO, Geneva, 2015.

Feige, Edgar L. “Professor Schneider’s Shadow Economy: What Do We Really Know? A Rejoinder.” Journal of Tax Administration, Vol. 2, No. 2, 2016.

World Bank. “Extending Pension Coverage to the Informal Sector in Africa.” World Bank Group, 2019.