Friedrich Schneider, Johannes Kepler University Linz

1. Introduction

The shadow economy includes economic activity deliberately concealed, partially recorded or omitted from official statistics. It matters because it weakens tax revenues and social insurance, distorts national income statistics, complicates labour-market regulation and reduces macroeconomic policy precision. Yet its size cannot be directly observed. Researchers infer it from indirect signals, so every estimate depends on assumptions about what is measured and how hidden activity leaves observable traces.

Between 2018 and 2025, this field became considerably more sophisticated. Digitalisation altered payment behaviour and weakened some traditional cash-based proxies, but at the same time created new sources of information. The COVID-19 pandemic showed that informality can respond rapidly to shocks, exposing the limitations of static models. Statistical agencies and researchers also gained better access to administrative records, energy accounts, tax-compliance data and platform-based indicators. As a result, the debate has shifted from finding one definitive estimate toward combining several imperfect but complementary sources of evidence.

This change is important because the shadow economy is not a single homogeneous phenomenon. Contemporary research distinguishes among underground production, informal production and illegal production, which differ in causes, statistical visibility and appropriate policy response. A domestic service paid off the books, undeclared self-employment, VAT fraud and illicit trade all fall within the broad universe of hidden activity, but they should not be expected to generate the same data signals. The central methodological challenge is therefore not simply to estimate "missing GDP", but to understand a latent economic system that leaves multiple and often noisy traces.

2. From Classical Methods To Hybrid Measurement

Earlier research relied mainly on four families of methods: national-accounts reconciliation, currency-demand models, MIMIC models and physical-input approaches. National-accounts methods compare the production, income and expenditure sides of GDP and incorporate survey and administrative adjustments. Their main advantage is consistency with official statistics, but they are data intensive, slow to update and dependent on the quality of statistical institutions. Currency-demand models infer hidden transactions from excess cash holdings. They offered long time series and straightforward implementation, but digital payments, precautionary cash demand, interest rates and financial instability have made the link between cash and informality less reliable.

Traditional MIMIC models represented a major conceptual advance by treating the shadow economy as a latent variable driven by multiple causes and reflected in several indicators. However, their estimates depend heavily on calibration, indicator selection and the assumption that one structural model can be applied across very different countries. Physical-input methods, especially those based on electricity consumption, faced a different problem: modern economies have become more service oriented, more energy efficient and more diverse in their energy use, reducing the reliability of electricity as a universal proxy for hidden output.

Recent innovations try to overcome these weaknesses rather than abandon the earlier approaches entirely. The Enhanced MIMIC framework developed by Asllani, Schneider and Dell'Anno introduces country-specific effects, dynamic corrections and more refined calibration. This allows the determinants of informality to vary across institutional settings. In high-income economies, taxation, labour costs and regulatory complexity may be especially relevant; in lower-income settings, weak enforcement, structural unemployment, governance gaps and limited state capacity often play a larger role. The method therefore improves comparative realism while retaining the multicausal logic of the MIMIC tradition.

A second important development is the hybrid National Accounts-Macroeconomic approach associated with Dell'Anno. Instead of treating official statistical adjustments and econometric inference as competing strategies, it combines national-accounts exhaustiveness procedures with macroeconomic residuals and labour-input discrepancies. This can reduce double counting, improve interpretability and distinguish more clearly among underground, informal and illegal components. In data-rich European settings, such hybrid methods appear particularly useful because they connect academic estimation with the practical needs of statistical offices.

The Total Energy Consumption method proposed by Achim and co-authors modernises the physical-input tradition by using aggregate final energy consumption rather than electricity alone. It therefore captures a broader range of production and is less affected by payment innovations. However, changing energy efficiency, decarbonisation and sectoral structure still weaken the stability of the energy-output relationship. VAT-gap and tax-gap estimates provide another useful complement: they do not measure the whole shadow economy, but offer administratively grounded evidence on the difference between theoretical tax liabilities and actual collections.

The common theme is methodological pluralism. Recent research increasingly recognises that different methods identify different margins of hidden activity. A VAT gap is not equivalent to hidden GDP, an energy discrepancy is not the same as undeclared labour, and a latent institutional model does not reproduce a national-accounts adjustment. The appropriate response is therefore triangulation, sensitivity analysis and explicit uncertainty reporting rather than artificial precision.

3. What The Latest Evidence Tells Us

Despite methodological differences, recent studies increasingly agree on broad comparative patterns. Advanced economies tend to have smaller shadow sectors than emerging and low-income countries, while institutional quality remains one of the strongest correlates of informality. Medina and Schneider's global estimates for 2010-2015 place the shadow economy at about 15.3% of GDP in OECD countries, compared with 33.4% in Latin America and the Caribbean and 36.2% in Sub-Saharan Africa. By income group, their estimates are roughly 15.0% in advanced economies, 27.6% in emerging markets and 38.8% in low-income countries.

European evidence points to gradual but incomplete formalisation. Kelmanson et al. find that informality declined across much of Europe after the global financial crisis but remained significant in Eastern Europe and parts of the periphery. Using total energy consumption, Achim et al. estimate average shadow-economy shares across 26 EU countries in 2001-2021 ranging from about 9.5% to 40% of official GDP. Lower values generally occur in more institutionally advanced and digitally integrated economies, while higher levels remain concentrated in parts of Southern and Eastern Europe.

A global update reported in 2025 suggests a decline from about 19.3% of GDP in 2000 to 11.8% in 2023 when countries are weighted by evolving GDP. Yet the arithmetic mean across countries is about 26.1% in 2023, showing how strongly conclusions depend on weighting conventions. The same evidence shows a stark income gradient, with around 5.9% of GDP in high-income economies and 42.4% in low-income countries. The long-run decline also slowed around major shocks, including the global financial crisis and the pandemic.

The more recent literature also moves beyond estimating size. Studies of developing countries find that larger shadow economies are associated with weaker economic growth, while financial inclusion can reduce informality and moderate its adverse growth effects. Access to bank accounts, mobile finance and digital payments lowers the transaction costs of formal participation and reduces reliance on opaque cash-based exchange. Digitalisation also changes observability itself: some activity becomes formal, some remains informal but leaves a clearer digital trace, and some migrates into new grey zones such as platform work, peer-to-peer commerce and crypto-based exchange.

4. Policy Implications And The Next Frontier

The policy message is clear: enforcement alone is insufficient. Formalisation combines credible oversight with lower compliance costs. Electronic invoicing, third-party reporting, interoperable business registries, simplified tax regimes and risk-based compliance can strengthen fiscal capacity without relying only on punishment. In labour markets, portable benefits, simpler contracts, digital wage reporting and transitional contribution schemes can make formalisation less costly for workers and firms.

Institutional legitimacy is equally important. Informality tends to be lower where public institutions are credible, corruption is limited, rules are perceived as fair and citizens view compliance as reciprocal rather than purely coercive. Tax morale and trust in government therefore belong at the centre of the analysis. Financial inclusion and digital public infrastructure also function as formalisation infrastructure: low-cost accounts, e-payments, e-ID systems and machine-readable invoicing reduce both the practical and administrative barriers to entering the formal economy.

The next generation of measurement will increasingly rely on high-frequency and data-rich sources. Web-scraped business listings, online marketplaces, mobile payments, satellite imagery and anomaly-detection tools can support nowcasting and provide local detail that annual estimates often miss. Yet predictive power is not the same as conceptual validity. These tools raise questions of privacy, proportionality, legal oversight and transparency. Similar caution applies to crypto-assets, which can facilitate hidden transactions but also generate permanent pseudonymous traces that differ fundamentally from cash.

5. Conclusions

The main lesson from the recent literature is that the study of the shadow economy has moved from methodological competition toward structured triangulation. Enhanced MIMIC models improve the treatment of heterogeneity, hybrid national-accounts approaches connect econometrics with official measurement, broader energy methods modernise physical proxies, and tax-gap and digital indicators add policy-relevant evidence. None is sufficient on its own.

Empirically, the evidence points to gradual reductions in many advanced economies but persistent and structurally embedded informality in many developing countries. Differences in taxation, regulation, governance, labour-market institutions, financial development and social norms mean that no universal explanation or policy package will fit every country. The future challenge is therefore twofold: to measure hidden activity more accurately while avoiding false precision, and to design formalisation policies that improve compliance without worsening exclusion. In this sense, shadow-economy research is becoming less a search for a single hidden number and more an inquiry into how modern states observe, regulate and integrate economic activity at the shifting boundary between formality and informality.

References

Achim, M. V., Postea, M. M., & Nojab, G. G. (2024). New estimate of shadow economy based on the total energy consumption: Evidence from the European Union countries. Energy Economics, 130, 107335.

Asllani, A., Dell'Anno, R., & Schneider, F. (2026). Global estimates of informal economies and new insights for 152 countries (1997-2022): Using an enhanced modelling approach. Economic Analysis and Policy, 89, 721-741.

Asllani, A., & Schneider, F. (2025). A review of the driving forces of the informal economy and policy measures for mitigation: An analysis of six EU countries. International Tax and Public Finance, 32(1), 310-344.

Dell'Anno, R. (2025). Integrating national accounting and macroeconomic approaches to estimate the underground, informal, and illegal economy in European countries. International Tax and Public Finance, 32, 526-565.

Dybka, P., Olesinski, B., Rozkrut, M., & Toroj, A. (2023). Measuring the model uncertainty of shadow economy estimates. International Tax and Public Finance, 30, 1069-1106.

Fu, J., Ahmad, W., Hussain, B., Khalid, M. W., Lambarraa-Lehnhardt, F., & Umair, M. (2025). Shadow economy, financial inclusion and economic growth nexus: Evidence from developing countries. Humanities and Social Sciences Communications, 12, 1724.

Kelmanson, B., Kirabaeva, K., Medina, L., Mircheva, B., & Weiss, J. (2019). Explaining the shadow economy in Europe: Size, causes and policy options. IMF Working Paper No. 19/278.

Medina, L., & Schneider, F. (2018). Shadow economies around the world: What did we learn over the last 20 years? IMF Working Paper 18/17.

Schneider, F. (2022). New COVID-related results for estimating the shadow economy in the global economy in 2021 and 2022. International Economics and Economic Policy, 19(2), 299-313.

Schneider, F. (2023). Do different estimation methods lead to implausible differences in the size of non-observed or shadow economies? A preliminary answer. Annual Review of Resource Economics, 15(1), 257-277.

UNDP Accelerator Labs. (2023). Informal businesses and the shift to digital: What we learned from small enterprises joining the digital economy.