Responsibility for opinions expressed in signed articles rests solely with their authors, and publication does not constitute an endorsement by the ILO.
This article is also available in French, in Revue internationale du Travail 165 (2), and Spanish, in Revista Internacional del Trabajo 145 (2).
1. Introduction
Recent years have seen a mushrooming of policy and academic accounts of the causes and consequences of large informal sectors in lower-income countries. In particular, a substantial macro comparative literature has emerged that regresses informal sector sizes against a range of variables and makes recommendations on social and economic policy (Williams 2015; La Porta and Shleifer 2008). This wave of scholarship has been facilitated by the increasing availability of country-level measurements of the size of informal sectors. These measurements have contributed to bringing together a field of study that is otherwise characterized by disciplinary and methodological diversity, and to feeding an evolving policy discussion. In 2017, the “proportion of informal employment in non-agriculture employment” became one of the indicators of the Sustainable Development Goals, exemplifying the targeting of the aggregate size of informality as an explicit policy goal.1 In 2021 and 2022, respectively, both the International Monetary Fund (IMF) and the World Bank published flagship reports on the developmental implications of informality, drawing from a variety of these measurements (Deléchat and Medina 2021; Ohnsorge and Yu 2022). Overall, these developments have created an impression of the informal sector as a clearly conceptualized, delineated and operationalized object of study that can produce clear goals and recommendations for policymaking. This article argues that this impression has largely been a mirage. It suggests that the way in which informal sectors have been measured over the first two decades of the 21st century has actively masked, and contributed to, continuing confusion in our understandings of what the informal sector is – and what its causes and consequences actually are.
Articles detailing the difficulties in defining informal sectors are as old as the concept itself and are a hallmark of its scholarship. The concept of informality emerged from attempts to describe and measure economies in lower-income countries that did not align with pre-existing concepts and assumptions. It was appropriated by various groups with varied interests, with different scholars and organizations emphasizing different aspects of informality (Benanav 2019; Peattie 1987). While informality was originally conceived in relation to its productivity or modes of production, contemporary approaches have often varied between defining informality in relation to registration with state institutions, or in relation to a list of activities (street vending, hawking) or common features (small scale, untaxed). Unsurprisingly, these do not always neatly align.
This article does not seek to expand on this heterogeneity but rather to examine how measurements of informality have proliferated despite these conceptual challenges and to what effect. It traces the evolution and application of direct and indirect measurements of informal economies, showing how in each case these measurements have moved conceptual decisions about the meaning of informality “downstream” into the specifications of statistical models and the operationalizations of the measurements. Consequently, the differences between common measurements currently in use are not primarily the result of inherent and understandable difficulties in measuring informality but of measuring fundamentally different things. The way in which these measurements have been integrated uncritically into policy and academic scholarship has obscured these differences, while what is actually being measured has often lacked transparency.
The article’s second argument is that this confusion has tangible consequences “upstream” – for our understanding of the causes and consequences of informal sectors and for the policy recommendations that are based on them. While at best there has been a seeming consensus that overlooks some spurious foundations, at worst data generation and a substantial part of recent scholarship on informality has been actively misleading, shaped by feedback loops of preconceptions and ideological priors. Meanwhile, findings on the supposed causes and consequences of informal sectors have substantial practical consequences for policymaking in a variety of fields, ranging from social protection to labour market regulation, taxation, poverty reduction and growth. Many prominent arguments remain sensitive to the measurement of the informal sector.
For example, a central part of the discussion on the costs of social protection policies has been whether they “increase informality” – a discussion that is highly sensitive to aggregate measurements of informal sectors (Calligaro and Cetrangolo 2023). Similarly, there has been policy enthusiasm in lower-income countries in recent years around mass tax registration and taxing mobile money, explicitly motivated by the idea that there are large informal sectors that are not paying any taxes (Moore 2023). However, as this article highlights, this idea is similarly dependent on the conception of informality itself. It can be true by assumption (in many indirect models), potentially true, or largely not true at all (in many operationalizations of direct measurements). And yet, the sensitivity to measurement issues is not usually addressed in the academic and policy literature on these topics. There are clear stakes: discussions on the causes and consequences of informality inform guidance on what kind of social protection systems States should adopt and where revenue authorities should focus their efforts. Parallel discussions exist around formalization incentives, direct cash transfers and urban planning. Critically, all these discussions are shaped by strong ideological priors – and, as this article points out, are characterized by measurements that are vulnerable to replicating them. While this article focuses on measurements of informal sectors, it also seeks to provide a wider case study of the interactions between complex measurements and the interests of broader policy narratives.
Following a brief review of the origins of informality, this article focuses on two measurements: the collection of direct measurements assembled and standardized by the ILO, and the school of indirect measurements afforded by multiple-indicators, multiple-causes (MIMIC) models. Together, these represent the dominant measurements of the informal sector used today. The article shows that they are rooted in two different conceptions of informality that have existed throughout its intellectual evolution – one based on a list of broad characteristics and the other on firms’ relationship with the State. Over recent decades, they have had contrasting trajectories. Direct measurements have evolved towards an increasingly precise operationalization that provides reliable estimates but is anchored in country-level characteristics and hence needs to be contextualized for comparative analysis. Meanwhile, the operationalization of indirect measurements has become increasingly obscure, opening substantial space for circular and ideologically pre-determined arguments.
As this article focuses on the origins of the data that have shaped the discourse on the informal sector up to now, it is naturally backward-looking. It does not propose or advocate for a specific definition of the informal sector. However, its conclusions, summarized in the final section, are highly critical of dominant indirect measurements and are largely in line with the recommendations and operationalizations of the 21st International Conference of Labour Statisticians (ICLS) resolution concerning statistics on the informal economy adopted in 2023. Despite its challenges, there is practical value in continuing to use and operationalize the informal sector – and in maintaining its increasingly common operationalization with respect to its relationships with the State. However, to make these measurements meaningful for comparative analysis and policy recommendations, transparency about the respective country-level institutions that firms interact with is crucial. In addition, registration-based approaches should be operationalized in a way that supports the core objectives of the policy discussion on the development of informal firms – namely, to improve access to social protection, increase productivity and promote decent work.
The specific focus of this article is the informal sector. While this is sometimes used interchangeably with the informal economy, they are substantially different: the latter is broader and includes “off the books” employment in formal firms, undeclared wages and hours in the context of formal jobs, and unpaid activities outside employment such as own-use household production. There are active and related discussions on the measurement of the informal economy and informal employment as well, both of which draw on sources of measurements similar to the ones covered in this article. However, the discussions on informal employment cover a range of additional aspects, such as the nature of employment contracts and social security provision, that lie beyond the immediate scope of this article. All of this is further complicated by the fact that some of the literature on indirect models tends to use some of these terms, along with the term “shadow economy”, interchangeably. This article’s discussion of the challenges posed by indirect measurements applies more broadly.
2. The issue of defining features
A striking feature of studies on the informal sector is that the term came first, with subsequent decades being increasingly focused on pinning down its definition. The modern popularization of the term can be traced back to the work of two anthropologists: Boeke’s (1953) work on Indonesia, subsequently taken up by Lewis (1954), and Harris and Todaro (1970); and Hart’s (1973) findings on Ghana, which were first presented in 1971 and subsequently taken up in a report by the ILO (1972). What both have in common is that they provide a description rather than a definition of the sector. This often included a list of occupations – early accounts frequently outlined common features of informal operators (ease of entry, family ownership, small scale), without identifying any of them as defining features.
The years that followed saw a rapid adoption of the concept, satisfying a demand for a language that could describe both the labour markets of many lower-income countries and their incongruence with conceptions of employment and unemployment that were dominating development discourse. Consequently, despite the definitional ambiguities, the idea of the informal sector was almost instantly entangled with a measurement project.
As Benanav (2019) has chronicled, both the take-up of the concept and its measurement have been intimately connected to the work of the ILO, which particularly through the ICLS, has continually codified, operationalized and refined its definition of the informal sector. Discussions in the ILO in the 1980s and 1990s already foresaw many of the challenges in defining and measuring informality. Measurement required clear operationalization, and any one of the various criteria commonly attributed to informality would produce a remarkably different measurement. While the 1980s and 1990s also saw registration with the State emerge as a potential defining feature, this presented similar challenges: there were many different kinds of registration, with small enterprises straddling various levels of state recognition (ILO 1992; Benanav 2019, 123). At the same time, the scope of the sector further broadened – having originally covered own-account workers primarily, it now also included “micro enterprises” of informal employers that had multiple employees. As Benanav (2019, 124) notes, this effectively hardwired heterogeneity into the concept of the informal sector, which came to encompass both survivalist activities and more growth-oriented and higher-income operators. This diversity underlies many of the contemporary challenges encountered in describing the sector and situating it in broader development discourse, and partially explains what are commonly described as the “four schools” of thought on informality (Chen 2012).
Subsequent studies and ICLS resolutions distinguished between the informal sector and the broader category of informal employment, recognizing that informal employment can also be found in households and “off the books” in the formal sector. There was also an increasing trend towards definitions and descriptions of the informal sector that emphasized registration and connection to the State. This is perhaps attributable to the difficulties in operationalizing features that focused more on the internal workings of informal activities, or waning interest in surplus labour and productivity, which had originally given rise to broader conceptions. Perhaps more importantly, this coincided with a range of parallel, albeit heterogeneous, policy trends, all of which emphasized registration: enthusiasm around titling (de Soto 1989) and recognizing informal livelihoods and organizations (WIEGO 2020), and increasing mainstream policy interest in firm formalization (Gallien and Boogaard 2023). While there are fundamental differences between these approaches, what they share is a view of formalization that is connected to the State and, consequently, a view of informality that is closely related to forms of registration.
While a large section of the scholarship on informality has closely followed the ICLS conceptions of the informal sector, there have also been two broader trends. One is the rise of concepts that effectively serve as substitutes for “informal sector”, and are at times used interchangeably, but do not generally follow a coordinated definition or measurement. These have included “the ingenuity economy, the economy of improvisation and self-reliance, the do-it-yourself, or DIY, economy” (Neuwirth 2011, 17–18), alongside increasingly common analogies with the gig economy, and, most prominently, the “shadow economy” (Schneider 2004). The second trend is the continued scepticism about descriptions of the informal sector as a binary, which has given rise to increasing references to it as a “spectrum” (Guha-Khasnobis, Kanbur and Ostrom 2006) and considerations of an index-based approach (Holland and Hummel 2022). One of the challenges of these approaches, however, is that they are at odds with a policy discussion that has become increasingly reliant on country-level percentage measurements.
3. Measuring the informal sector
Given the difficulties involved in measuring informality, it is unsurprising that there has been a wide range of methods used to measure the informal sector, ranging from using labour and household surveys to analysing currency demand or electricity to (rather surprisingly) surveying leaders of formal firms about the size of informal sectors (La Porta and Shleifer 2008). Meanwhile, two methods have become increasingly dominant. The first – the data assembled by the ILO based on enterprise, household and mix surveys (ILO 2018) – relies on direct measurements. The second – a variety of estimates all based on MIMIC models – relies on indirect modelling. It is notable that they typically cannot be compared directly – one measures the informal sector as a proportion of the labour force, while the other measures it as proportion of gross domestic product (GDP). Both have contributed to the increasing availability of cross-country comparable data on informal sectors, aiding the rapid emergence of studies on their causes and consequences. There have been substantial methodological discussions about both types of measurements in recent years (ILO 2019; Breusch 2016; Kirchgässner 2017; Feige 2016). The goal here is not to review them comprehensively, but to examine how these methods address the ambiguities in the conceptualization and definition of informality, and to what effect.
I argue that in the face of remaining conceptual ambiguities, both approaches have de facto moved decisions on the nature of informal sectors “downstream” into technical specifications of their measurement processes. Subsequently, this appears to be poorly understood by many researchers and policymakers, who have frequently used these measurements in a way that lacks both transparency and specificity, generating the mirage of comparative measurements, while frequently measuring and comparing fundamentally different things, generating spurious findings and recommendations.
3.1 Direct measurements
The primary challenge in obtaining direct measurements of informal sectors for comparative analysis is establishing a definition that can be operationalized across survey data and national statistical offices in different contexts. Since the recent popularization of the concept of the informal sector, the ILO and ICLS have been at the core of this endeavour, both in shaping the operationalization and the compilation and analysis of data. The ILO has become the go-to resource for direct measurements of the size of informal sectors.
Closely embedded in a community of learning and practice, the ILO’s approach has been iterative – operationalizations have been refined through ICLS resolutions, and its compilation of data expanded to over 100 countries in 2018 (ILO 2018). The main purpose of Women and Men in the Informal Economy: A Statistical Picture (ILO 2018) (hereafter A Statistical Picture) was to measure informal employment rather than the informal sector, although it contains data on both. While it acknowledges that the quality of its data is dependent on the surveys conducted by national statistical agencies, and that there is “a certain degree of flexibility” (2018, 6) in the measurements taken, it is explicitly geared towards comparability.
While a comprehensive summary of the ILO’s measurement of the informal sector, which has already been well documented (ILO 2019, 2013; Benanav 2019; Hussmanns 2005), is beyond the scope of this article, it is worth focusing on two resources: A Statistical Picture and the 2023 ICLS resolution.
3.1.1 A Statistical Picture
A Statistical Picture represents, to date, the most comprehensive assemblage of country-level measurements of the size and composition of informal sectors based on direct measurements.2 It is frequently cited and is the basis for other statistical overviews (Bonnet, Vanek and Chen 2019). It has remained the dominant reference for academic and policy literature on informal sectors since its publication. Consequently, even though the ILO’s definition and operationalization of the informal sector has since been updated, it is worth discussing A Statistical Picture.
A Statistical Picture also illustrates the challenges of defining and operationalizing informality in the context of competing conceptions. Its data were collected through an operationalization of the 15th ICLS definition of the informal sector. Mirroring the aforementioned tension between the registration versus common features approaches, the years leading up to the 15th ICLS had seen considerable disagreement on whether informality should be characterized by the internal features of an enterprise or by its relationship to the relevant legal and administrative framework. The resulting definition was thus a hybrid – there “was no agreement at the 15th ICLS as to which of the two approaches was preferable. The definition in the 15th ICLS resolution therefore incorporated both approaches” (ILO 2013, 18).3
The general definition did not mention registration and focused on common, rather than defining, features:
The informal sector may be broadly characterized as consisting of units engaged in the production of goods or services with the primary objective of generating employment and incomes to the persons concerned. These units typically operate at a low level of organization, with little or no division between labour and capital as factors of production and on a small scale. Labour relations – where they exist – are based mostly on casual employment, kinship or personal and social relations rather than contractual arrangements with formal guarantees. (ILO 1993, 52)
Registration, however, came into its operational definition. It offered two different conditions under which enterprises could be classified as informal. One was based on registration status, while the other was based on the number of employees, presumably as a proxy for “modes of production” conceptions of informality. The latter left substantial scope for how each proxy should be interpreted. Registration “may refer to registration under factories or commercial acts, tax or social security laws, professional groups’ regulatory acts, or similar acts, laws, or regulations established by national legislative bodies” (ILO 1993, 53). The size limit “may vary between countries and branches of economic activity. It may be determined on the basis of minimum size requirements as embodied in relevant national legislations, where they exist, or in terms of empirically determined norms” (ILO 1993, 54). Consequently, the definition resolved the tensions between different conceptions of the informal sector by moving them downstream into its operationalization and essentially maintaining both as options.
A Statistical Picture introduces one further refinement to this operationalization: it establishes a hierarchy between registration and size. Regulation comes first: for countries in which data on registration are available, this becomes the primary variable – with size as a criterion of “last resort” in the absence of any registration data. Notably, while the operationalization establishes a clear cut-off in terms of firm size (five or fewer employees), it does not provide any clarification of the type of registration under consideration. It notes that “this includes registration with social security authorities, sales or income tax authorities and should be at national level” (ILO 2018, 8). This still contains a huge range of different types of registration.
Perhaps most perplexingly, the publication does not distinguish between measurements that have been obtained through firm size and registration. It specifies that the breakdown between the two approaches is almost even: for 48 per cent of the countries studied, the “alternative” approach has been taken (ILO 2018, 83). De facto, for half of the reporting countries, the reported size of the informal sector represents a measurement of the number of unregistered enterprises, while for the other half, it represents a measurement of the number of enterprises with under six employees. But it is unclear which approach has been used in which country.
Thus the old division between registration- and organization-focused conceptions of the informal sector is decided through data availability. Part of this is a technical issue, and an entirely reasonable operational approach given the lack of available data. One of the greatest assets of A Statistical Picture is its awareness and management of the limitations of surveys. But due to the lack of transparency on which approach has been used for which country, this technical issue becomes conflated with a conceptual one, as data availability is used to decide between two different conceptions of informality.
3.1.2 An illustration: Informality in Accra
It is useful to illustrate that these approaches are not only conceptually, but also empirically different. For the difference between size and registration, this is self-evident. As ILO analysis has highlighted, firm size is a highly imperfect predictor of registration status and vice versa. For example, according to a 2018 household survey in Argentina, most firms with between 5 and 49 employees were not registered. Meanwhile, according to a 2018 labour force survey in Rwanda, less than 50 per cent of firms of all sizes were registered, but more than 30 per cent of firms with 5 or fewer employees were registered (ILO 2019).
These issues persist with respect to the type of registration. In order to illustrate this point, I draw on survey data from 2,700 informal enterprises that are a representative sample of informal enterprises in urban Accra, Ghana, collected through a joint ICTD-WIEGO-ISSER research project.4 These 2,700 enterprises are determined to be informal by the registration proxy that is used by the statistical service in Ghana and would fall under any ILO definition of informality since the 15th ICLS, as they are not registered with the Registrar General. Most of these enterprises are also small – over 80 per cent consist of own-account workers and, for those of them with paid employees, the vast majority have under 5 employees. It should be noted, however, that small firm size is a common feature of Accra’s economy more broadly (Teal 2023).
The survey data allow us to simulate how the size and composition of Accra’s informal sector would change if we refrained from applying the “registration with the Registrar General” conception of registration and instead applied other forms of registration. For example, 8.8 per cent of this sample report being registered with the Ghana Revenue Authority (GRA) for tax purposes, while 22.5 per cent report making regular tax payments to the GRA. Consequently, shifting the form of registration used towards a tax-based registration proxy would have a substantial impact on the size of the informal economy in Accra. The 15th ICLS operationalization also allowed registration with social security authorities; 62 per cent of the respondents in this sample (meaning own-account workers or employers) are registered with the National Health Insurance Scheme (NHIS) set up in 2003. If this were to be used as the relevant form of registration, the size of Accra’s informal economy would suddenly shrink dramatically.
Moreover, while there is general consensus that registration should be at the national level, looking at local registration shows that this overlooks a huge proportion of the informal sector in Accra: 40.4 per cent of informal enterprises report being registered with the Accra Metropolitan Assembly (AMA), while 59 per cent report paying some form of taxes to the AMA, highlighting that the focus on national registration is overlooks a significant share of registration more broadly.
Changing the type of registration not only changes the size of the informal sector – it also changes its features. For example, when we look at registration with the Registrar General, the median monthly gross income of informal enterprises in Accra is 900 cedis (approximately 150 US dollars at the time of data collection). If we use registration with the GRA instead, that median income shrinks – not hugely, but noticeably – to about 868 cedis. The proportion of informal enterprises that report having access to a bank account similarly falls, as does the proportion of informal employers. Using registration with the NHIS substantially shifts the gender balance of the informal sector, as women are more likely to be registered with the NHIS, and so on.
Accra also provides an important reminder that government registration strategies are changing and that new forms of registration and connections between citizens, enterprises and States are emerging. Some 78 per cent of informal workers in Accra report that they have a Ghana Card, which is a national ID card. In 2021, the GRA announced that the Ghana Card personal identification number would replace the Tax Identification Number, the Social Security and National Insurance Trust biometric number and the National Health Insurance number. Under this reform, formality defined as registration with one of these bodies would suddenly increase substantially, with little change in the economic reality of the people involved. I will return to this point in the final section.
What this empirical aside highlights is that the broad scope of potential levers of registration remains a substantial problem for comparability of the size of informal sectors across countries. The choice of registration affects both the size and the substantive features of the sector, and consequently the very basis for comparison. It also highlights the fact that, as new forms of registration emerge, and the nature of work and relationships with the State itself change, central assumptions that these types of registrations broadly overlap do not necessarily hold. As the next section highlights, this fact has been at the centre of the most recent ILCS work on this issue.
3.1.3 The 2023 ILCS
The ILO’s conceptualization has evolved since 2018’s A Statistical Picture, partially in response to some of the issues raised above. The 21st ILCS resolution, adopted in 2023, introduces various adjustments to the conceptualization and measurement of the informal sector, including the inclusion of agriculture in the informal sector and an increase in the threshold for the proportion of goods that must be intended for the market in an informal enterprise. What is most relevant to the discussion here, however, is its resolution of the registration issue.
Here, the new operationalization formalizes the prioritization of registration that was previously implicit – firm size is “no longer considered a main criterion” (ILO 2024, 11). Furthermore, the resolution provides additional detail on what kind of registration should be considered. It notes that:
Registration should refer to a register or registers in the given country used for granting access to benefits such as tax deductions, obtaining a separate legal identity for enterprises, granting access to statutory social insurance (if it implies a formal status of the economic unit) and carrying obligations such as paying business tax and keeping accounts. The register or registers would typically be at a national level, but could also be at a local level if the register is governmentally established and controlled, but locally administrated. (ILO 2023b, paragraph 33)
It also notes that:
In countries where registering an enterprise might not necessarily carry any obligations or benefits, there might be a need to combine different registers such as, for example, the business register and the tax register, to ensure that a certain degree of formal arrangements comes with the formal status of the economic unit. (ILO 2023b, paragraph 34)
This resolves the primary issue with A Statistical Picture – namely, the use of size in the absence of registration data, and the lack of transparency around which measure has been used – and provides an interesting starting point for what is now effectively the most relevant discussion on the delineation of the informal sector: the appropriate type of registration. The focus on obligations and benefits, with the acknowledgement that these may not always be found at the same level of registration, resolves the issues created by very limited forms of registration (such as the use of national ID numbers as tax identifier numbers) that could otherwise “formalize” vast numbers of informal operators without providing a real change in their relationships with state structures. There is a clear theoretical coherence here, too, between this area of focus and a wider conversation on which government relationships are most meaningful for small firms.
The logical test and challenge lie in the data collection. Hopefully, the coming years will see the inclusion of registration and account-keeping variables in countries where these were previously not available, including through surveys that ask dependent workers relevant questions, in order to improve measurements based on the new operationalization. At the same time, the transparency issue still needs to be resolved – the centrality of registration in the new approach clearly merits a detailed analysis of the type of registration used. A conversation about the role of active and passive registration is likely to emerge. For example, recent years have seen many revenue authorities conduct mass tax registration exercises or use third-party data to mass assign tax identification numbers. Research has shown that the vast majority of newly “registered” firms often do not actually file or pay taxes (Mascagni et al. 2022; Gallien et al. 2023), highlighting that even though obligations may be implied through registration, it remains unclear how tangible these obligations are. The same, of course, applies to benefits. I will return to this point in the final section.
3.2 Indirect measurements
Direct measurements of informal sectors are reliant on the surveys compiled by national statistical agencies. Consequently, they generally cannot produce the kind of country-year informality panel data that are of particular interest for macro comparative econometric analysis of informality. Such data have been obtained through indirect measurements, which explains their enormous popularity in recent years. Here, the work of Friedrich Schneider and his colleagues using MIMIC models and Elgin’s (2021) more recent variations have been dominant.5 Cheap to produce and based on publicly available data, work based on this method has given rise to a huge swath of publications, focusing in particular on the “causes and consequences” of informality.6 Indirect measurements of informality have become predominant in recent years in World Bank and IMF flagship publications (Ohnsorge and Yu 2022; Deléchat and Medina 2021).
Built on a methodology first used by Frey and Weck-Hannemann (1983), MIMIC models treat the size of the informal sector as an unobserved latent variable. To estimate size, they require a theoretical model that posits causal relationships between time-varying causes and indicators. This is usually public data, available yearly and across countries. There has been huge variety in the respective specifications across the literature, with common causes including tax burdens and tax morale, business freedom and other indices of regulatory burdens, and common indicators including employment and GDP change. Notably, MIMIC models usually do not estimate the size of informal sectors as a percentage of the labour force, but instead as a percentage of GDP.
The expansion of the use of MIMIC models to estimate the size of the informal sector has been accompanied by a range of methodological critiques, many of which have not only found it fundamentally unsuitable for the study of informal economies, but also severely criticized the lack of professional scientific standards in its application.7 I do not summarize them comprehensively here, but instead focus on examining how these models have related to the conception of the informal sector.
Two points of critique, however, are particularly relevant. The first is MIMIC’s reliance on benchmarking. As MIMIC models generate an index variable rather than a total size of the informal sector, they need to be benchmarked against a measurement of the size of an informal economy that has been estimated using a different methodology. Through calibrating the “level” of the data, this benchmark has a huge impact on the measurements obtained. In this regard, indirect models in recent years have somewhat taken on the appearance of a house of cards, with increasingly complex models still relying on their predecessor’s foundations. Elgin’s (2021) dynamic general equilibrium (DGE) model, for example, is benchmarked against a previous version of a MIMIC model developed by Friedrich Schneider (Elgin 2021). Schneider’s models also require benchmarking, typically against a currency demand model. Currency demand models themselves also need to be benchmarked: common practice is to assume a “base year” in which informality is zero or close to zero – a rather remarkable assumption given everything we know about the history of informality (Kirchgässner 2017, 6; Medina and Schneider 2021, 24). With different models drawing on other models with increasingly weaker empirical standing and stronger assumptions, a “zero base year” assumption represents an extremely shaky foundation for this house of cards – or in the words of one of its critics, “numerical accidents without connection to the data” (Breusch 2005, 387).
3.2.1 Operationalizing without a definition
The second critique, which is more relevant to the focus of this article, is the issue of how indirect measurements affect the definition of the informal sector. Curiously, the literature about indirect measurements generally has been less concerned with definitional questions than the literature focusing on direct measurements. Skipping directly to the data, a notable amount of scholarship in this field provides no definitions, provides multiple definitions or simply expresses the perception that definitions of this topic are “difficult” (Elgin 2021, 7). While Schneider has provided some definitions in his more recent works, his earlier models were accompanied by the idea that “in general, a precise definition seems quite difficult if not impossible” (Schneider and Enste 2000, 79; Feige 2016). This literature sees a variety of different terms used inconsistently or de facto interchangeably, even though they are not considered interchangeable in the literature on direct measurements. These terms include “informal sector”, “informal economy” and “shadow economy”.
Underneath all this conceptual ambiguity, indirect measurements have their own operationalized definition “downstream” in their methodology. While the operationalizations of the direct measurements outline which enterprises are considered informal, theorized causes and indicators do the same for the MIMIC models. They, in essence, describe what aspects of the economy are used to estimate informality – i.e. what characterizes informality. This explains the somewhat relaxed attitude of this literature to definitions – essentially, it goes back to the very early literature on the informal sector, not through a definition, but through a list of features. The MIMIC model thus presents a different methodological solution to the absence of a unified definition of the informal sector. If the ILO’s measurements were previously based on the operationalization of a hybrid definition, the MIMIC methodology operationalizes the absence of a definition. While it provides a neat methodological parallel to conceptions of informality that rely on a list of features, this is precisely where the issues with this solution become particularly stark.
Perhaps the downside of model-based approaches to measuring the informal sector is their reliance on their specifications. Their estimations are dependent on the choice of causes and indicators. With its origins in psychometrics, MIMIC is designed for a context in which we have a very good sense of the causes and indicators of a phenomenon that cannot be observed itself. It is highly questionable whether this is true for informal economies – MIMIC is, as its proponents point out, a confirmatory rather than an exploratory model. This then puts substantial pressure on the specification of what is being confirmed – particularly on the selection of indicators.
Crucially, there is no consensus on the selection of causes. They are often “somewhat arbitrary” (Elgin 2021, 24) and vary substantially in number and size. There has been substantial critique of some of the chosen indicators – Williams’ review of the method labelled some of the common causes and indicators as “highly questionable”, noting in particular the use of tax rates (Williams 2023, 59). While some of the literature explicitly states the causes that have been included in the model, much of it, especially the literature drawing on pre-existing MIMIC datasets, is highly opaque regarding which factors have been included and why, and how they have been measured, with substantial sections of the literature not providing enough information for replication (Feige 2016). What is almost always left unsaid in MIMIC models is that the causes do not emerge from a theoretical consensus on what the informal sector is and what it is caused by – because such a consensus does not exist. Instead, the causes are either entirely arbitrary or themselves advance a theoretically informed conception of informality. I will return to this later.
A further complication of the “causes approach” of conceptualizing informality is methodological. By using a priori “causes” as a basis for measuring informality, the resulting measurements are inherently limited in their application, with two consequences. First, while MIMIC models assign a relative causal weight to the various predefined causes, this must not be confused with an actual causal explanation of informality. For example, one MIMIC model identifies the “driving forces of informality” to be personal income tax (13.8 per cent), indirect taxes (14.1 per cent), tax morale (14.5 per cent), unemployment (14.7 per cent), self-employment (14.5 per cent), GDP growth (14.3 per cent) and business freedom (4.3 per cent) (Elgin 2021, 37). However, these proportions are determined by the causes selected in the model set-up and cannot be taken as an indication of a causal relationship between these factors and the “real” informal sector. Unfortunately, such caution is not always applied.
The second consequence of the inclusion of theoretical causes in the construction of the model estimates of informality is that the models using these estimates cannot be used to test the impact of these same causes on informality. For example, if social security contributions or tax morale are among the causes used to construct an estimate of informality, this estimate cannot be used to test whether social security contributions or tax morale are drivers of informality. While this issue has been clearly acknowledged by the designers of these models, its incorrect application is still widespread (Kirchgässner 2017).
Despite their popularity, indirect approaches through MIMIC models have not resolved or overcome the absence of a clear definition of, or conceptual consensus on, the informal sector. Instead, they mask their position within a huge array of conceptions by pushing theoretical and conceptual decisions “downstream” into methodological specifications, where they are less visible but remain impactful. As the next section discusses, this may in fact be one of the very reasons why they have remained so popular, despite their methodological shortcomings.
3.2.2 An illustration: Measurements, politics and taxes
Scholarship on the informal sector has always been intimately tied to the question of its causes and consequences. This has expanded through the wider availability of statistical measurements of informality, especially through indirect methods. In 2021 and 2022, the World Bank and the IMF published flagship reports on informal employment and the informal sector that drew heavily and explicitly on new scholarship and the above-mentioned measurements to advance policy recommendations. While this interest in an evidence-based policy approach to informality is important, it also provides a case study of the dangers and consequences of the reliance on quantitative measurements in the absence of conceptual clarity. In particular, the reliance of indirect models on strong theoretical assumptions risks confirmation bias within research and policy writing, where previously held assumptions are fed into model design and the estimates of these models are then inappropriately interpreted causally. This contributes to a conflation in common discourse between defining and common features, as well as causes and consequences, of informality, which leads to misleading policy discussions.
The Two recent IMF (Deléchat and Medina 2021) and World Bank (Ohnsorge and Yu 2022) reports provide useful illustrations. Both acknowledge that a multitude of definitions of the informal sector exist, but do not present them as something that is fundamentally relevant to or that drives analyses and insights. Notably, they advocate for the use of multiple measurements, acknowledging that the differences between their estimates result from different measurement approaches.8 While both acknowledge that there is disagreement on how informality is measured, there appears to be no disagreement on what is being measured.
While both reports use MIMIC models, there is almost no discussion on how the specification of these models was chosen, or the implications. De facto, both models’ assumptions represent a highly voluntarist understanding of informality as driven by operator-level decisions to avoid burdensome regulation. The MIMIC model used in the IMF report includes tax burdens, regulatory burdens and the unemployment rate (assumed to increase informality), as well as an economic freedom and business freedom index (assumed to decrease it). The World Bank model includes the size of government, share of direct taxes, a fiscal freedom index and business freedom index (both developed by the Heritage Foundation), the unemployment rate and an indicator of government effectiveness.
These lists of assumed causes are not uncontroversial and, in contrast to the reports’ overall tone, do not represent a consensus understanding of informality. What these models consequently generate is really an index that relates changes in regulatory burdens to changes in employment and GDP over time. This would be less concerning if it was communicated transparently as a theoretical choice or accompanied by the utmost caution in how the resulting data are used and interpreted, especially with respect to causal relationships between regulatory burdens and informality. But this is seldom the case.
Similar examples can be found throughout the literature on tax and informality. Besley and Persson’s 2014 article in the Journal of Economic Perspectives entitled “Why Do Developing Countries Tax So Little?”, to date perhaps the most cited article in the entire literature on tax and development, presents a scatter plot of the size of the informal economy and the share of income tax in revenue. The article discusses their correlation and declares that “an increase in formality is a key part of the process by which taxation increases with development” (2014, 110). What it does not discuss are the origins of its estimation of informality – it merely cites an unpublished working paper by Friedrich Schneider from 2002. That paper does not provide details of the variables in its MIMIC model, but lists among the model inputs “the burden of direct and indirect taxation” because “a rising burden of taxation provides a strong incentive to work in the shadow economy” (Schneider 2002, 41). In the context of a lack of transparency around measurement specifications, pre-conceived assumptions, measurements and the interpretation of resulting correlations come full circle.
There is a similar lack of clarity in the World Bank report. While it is careful not to make causal claims, it makes a huge number of associational claims, essentially pointing out correlations. The phrase “associated with” appears 179 times in the report. While this is not in itself inappropriate, the phrase needs to be used with the utmost caution in a context where one of the key measurements has been obtained by associating informality with a range of variables, the relationship of which to informality is contested. The point here is not to nitpick these reports but to highlight that what is emerging is a feedback loop between assumptions, findings and recommendations, risking confirmation bias and the repetition of commonly held and ideologically informed, but not empirically verified, beliefs about informality.
This has been an explicit concern in critiques of MIMIC models for decades. In 2016, Breusch noted that the “complexity of the estimation procedure, together with its deficient documentation, leave the reader unaware of how these results have been […] shortened to fit the bed of prior belief”. Feige (2016, 22) similarly concluded that the user of MIMIC models may find “consistent substantive results conforming to his prior beliefs by selecting indicator variables and normalization coefficients that vary from study to study”. A decade later, this dynamic is not only still visible but has expanded into policy discussions.
Taxation provides an important example. Both reports draw on models that incorporate the common voluntarist position that informality is associated with lower tax revenues and is partially caused by tax burdens, and explicitly repeat it in their analysis. However, if we look beyond macro models, this is neither true by definition nor uncontroversial empirically. Indeed, while there may well be macro-level correlations between some tax measures and informality, substantial literature in recent years has highlighted that many informal enterprises do pay a range of taxes and fees, both to local and national levels of government (Carroll 2011; Anyidoho et al. 2025). At the same time, there has been a relative lack of micro-level evidence for the hypothesis that the costs of registration, and tax levels in particular, are key factors in keeping firms informal, with research instead pointing towards highly complex context- and sector-specific dynamics (Williams 2023, 35; Ulyssea 2020). The relationship between tax and informality is an ongoing and active area of research without a clear consensus. And yet, the wider association of large informal sectors with low tax revenue has remained a truism in the “causes and consequences of informality” discourse. Moreover, it has actively and directly contributed to policy enthusiasm around registering informal firms for taxes, which has largely been ineffective both from a revenue and equity perspective (Gallien et al. 2023; Moore 2023; Gallien and Boogaard 2023; Rogan 2019).
This hints at the consequences of the widespread use of indirect models of informality in a context where there is a lack of a clear agreement on how the informal sector is conceptualized. It not only contributes to confusion about the object of the discussion, but to an increasing reliance on its associations and correlations with wider macroeconomic features based on measurements, the details of which are not made transparent and are largely derived from these associations. This risks, for scholars and policy practitioners, confusing causes and consequences, as well as common associations and defining features, and creating feedback loops around previously held preconceptions of informality. Given that there are important dynamics within the informal sector that we do not yet fully grasp because of how much the sector is still changing, and given how much is at stake in the policy implications, this can lead to the perpetuation of bad policymaking that simultaneously references measurements and is increasingly resistant to independent evidence.
4. Implications: Making registration matter
The previous sections have argued that while new measurements have contributed to an extensive literature and a sense of iteratively accumulated knowledge on the informal sector, there remains substantial heterogeneity in how it is conceptualized. Key questions around what informality is have not disappeared but have been “downstreamed” into methodological specifications. Meanwhile, the way these measurements are presented and used has often failed to be transparent about the implications of their underlying specifications. One consequence has been that beyond very technical conversations on direct measurements, there is now a substantial amount of vagueness and confusion about what the informal sector actually is. Perhaps more gravely, this ambiguity, along with the way that indirect measurements have been used, has created a risk of feedback loops between assumptions and findings, and spurious analysis of the causes and consequences of informality. Where do we go from here? A few implications and consequences emerge from the argument made in this article.
First, as noted above, one of the most substantial issues in both direct and indirect measurements has been a lack of transparency. In studies relying on direct measures, such as future versions of the ILO’s A Statistical Picture building on the 2023 ICLS resolution, this requires specifying the precise criteria through which enterprises have been defined as formal or informal – including which form of registration has been used in each country. With respect to indirect methods, articles using any data generated by MIMIC models should provide both the full specifications of their models, including their benchmarking and a discussion of their consequences for the analysis. While transparency is needed at the level of data generation, the most critical issues here are found in the scholarship that uses existing data, and even more so in policy reports that draw on them. The use of multiple measurements concurrently is not an obvious solution – while it is often presented as ensuring rigour, it risks actively diluting what is being measured.
Second, what is currently presented as the comparative and quantitative “state of knowledge” on the causes and consequences of informality needs to be treated with the utmost caution. This is particularly true for the literature building on indirect methods. The risks of large sections of the perceived wisdom in this area being methodologically spurious, misunderstood or reinforced through feedback loops should spur a cautious rethinking of all common assumptions and relationships. Arguably, in the absence of comparative survey data based on the new ICLS operationalization, there are currently no data on the size of informal sectors that can be assumed to be reliable for cross-country comparative analysis without further specification of their mode of collection and analysis. One central task for scholarship in the coming years will be to re-examine propositions currently treated as established, in the light of new data. In the meantime, caution is warranted when advising policy.
Third, the most obvious resolution of some – though not all – of the issues raised in this article is the creation of a clear consensus on an operationalized definition. This is especially true as the contributions of indirect measurements to the academic and policy discussions provide a strong case for focusing on direct measurements going forward, for which clear definitions are essential. It is not the purpose of this article to advance one view or definition. But the issues raised here have implications for the process, function and context of such a definition.
First, there is still a strong case for establishing such a definition. Throughout the challenges of operationalizing informality, the suggestion to drop the term altogether is about as old as the term itself (Peattie 1987). While the term has substantial challenges, it has now become both deeply embedded in consequential policy discourses and a focal point for conversations and the mobilization of various actors, academics and activists. Clearly, clarity around the term would be more impactful than its abandonment.
Second, in order to be useful for purposes of measurement, the term needs clear boundaries. These do not need to be binary, but they need to be explicit. The observation that many features of interest regarding informality – such as economic vulnerability – exist on a spectrum or on both sides of any definition is important: the fact that there are different levels of risk and protection within and outside informal sectors is one of the key insights of the literature on this issue and needs to be a starting point for policymaking. But is not an argument against a clear definition.
Third, the definition needs to be transparent and explicit in how it conceptualizes the informal sector. Defining criteria need to be moved back “upstream”, from methodological specifications into definitions.
Fourth, there needs to be a process that allows definitions to adapt both to our expanding knowledge base on informality, and to how informal sectors themselves are changing in relation to technological change and state intervention.
Fifth, these definitions need to be practical and institutionally feasible. Measurement is a practical and logistical task – new methods need to build on the existing literature and measurement methods.
It should be clear from this list that the new ICLS definition is an obvious candidate that meets all of these criteria – and possibly the only candidate that meets the final two criteria. It provides an opportunity to generate cross-country measurements that are more comparable and more transparent than anything currently in use. It also formalizes a de facto trend in the conceptualization of the informal sector towards registration. In line with the arguments made in this article, it will be useful to be explicit about this focus on registration – to make it a more central part of how we speak about the measurements generated through this definition, and what they do and do not tell us. This includes highlighting the fact that the definition does tell us something about registration, but does not imply that informal firms are not registered sub-nationally, and that it may tell us something about tax registration, but not necessarily about payment or non-payment.
Most importantly, this definition formalizes a shift in the debate on the boundaries of the informal sector towards the question of what kind of registration matters. The 2023 ICLS definition has made a start here, pointing to both burdens and benefits resulting from registration. Its operationalization in different country contexts will likely add further substance to this discussion. Perhaps the most important consequence of this shift is that our conception of informal sectors is increasingly measuring something about the State – about its registration policies and capacities. It is likely that in many cases, this is increasingly developing into a function of the features and ambitions of the respective social protection system and national level tax registry. This in itself is valuable, but requires a shift away from a common framing that sees the State as primarily reactive to informality.
It also should provide further caution against making reducing the size of informal sectors in the abstract a policy goal in itself. When setting goals for the effectiveness of these policies, we should look at their performance on their own terms. This is particularly relevant in the face of a marked increase in the ways in which States register and interact with economic activities, ranging from new forms of digital IDs, simplified forms of registration, and new social protection schemes and cash transfers tied to registration programmes, some of which have accelerated as a result of the particular challenges faced by informal workers during the COVID-19 pandemic (Chen et al. 2022; Meagher 2022). With a more registration-focused definition, evaluating or justifying any of these policies with a reference to “reducing informality” is again circular.
Instead, this discussion on which forms of registration constitute formalization offers a chance to recentre why formalization is a meaningful policy goal in the first place. As the conception and measurement of informal sectors shift towards the State and away from informal sector operators’ economic experiences, these experiences must inform both the choice of the intersections with the State and their evaluations. This, again, requires recognition that formalization itself does not always address the economic vulnerability that is frequently found on both sides of any delineation of the informal sector, that formalization interventions are best conceptualized with a broader view of the economic experiences and livelihoods of informal operators, and that informalization within the formal sector also presents a serious challenge. Conveniently, this is precisely where the strength of decades of research and advocacy on informal sectors and on formalization lies – and particularly the qualitative, ethnographic and participatory work that has not always been included in some of the discussions, which have been built primarily on country-level comparative measurements.
Notes
- United Nations General Assembly Resolution A/RES/71/313. ⮭
- An update was published in 2023 (ILO 2023a) containing updated data across a huge range of indicators, including breakdowns by region. However, as it does not contain the same extensive country-level appendix as 2018’s A Statistical Picture, the latter remains the go-to resource for these data. ⮭
- See also Hussmanns (2001, 2005). ⮭
- For more information, see Anyidoho et al. (2025). ⮭
- I do not discuss dynamic general equilibrium (DGE) models here. Some of the issues discussed here apply to DGE models, while others are more complex. DGE does provide more micro-foundations, but is still vulnerable to the assumptions embedded in these foundations. ⮭
- See Feige (2016, 25–26) for a list of key variables analysed using this method. ⮭
- For summaries of these critiques, see Kirchgässner (2017), Breusch (2016) and Feige (2016). ⮭
- While chapters in both reports are written by different authors, there is no discussion of how conceptions of informality in these chapters relate to each other. ⮭
Acknowledgements
I would like to thank Niken Wulan and Josedomingo Pimentel Cavalié for their research assistance, Mike Rogan, Vanessa van den Boogaard, Michael Frosch and Rodrigo Negrete for their insightful comments on earlier versions of this article (including IDS Working Paper 599), as well as three anonymous reviewers and the journal editors for their comments and corrections. I would also like to thank the organizers and participants of the panel “Toward a coherent understanding of the crisis in the world of work: Centring social reproduction and informality in the pandemic age” at the 2023 annual meeting of the Development Studies Association. I would like to thank the International Centre for Tax and Development for funding this work, which has also been funded by UK International Development from the UK Government under grant number 300211-101, the Gates Foundation under grant number PP1197757 and the Norwegian Agency for Development Cooperation under grant number QZA-17/0153. The findings and conclusions contained within are those of the author and do not necessarily reflect positions of the funders.
Competing interests
The author declares that they have no competing interests.
References
Anyidoho, Nana Akua, Max Gallien, Michael Rogan, and Vanessa van den Boogaard. 2025. “The Taxed Informal Economy: Fiscal Burdens and Inequality in Accra”. World Development 187 (March): Article No. 106879. http://doi.org/10.1016/j.worlddev.2024.106879.
Benanav, Aaron. 2019. “The Origins of Informality: The ILO at the Limit of the Concept of Unemployment”. Journal of Global History 14 (1): 107–125. http://doi.org/10.1017/S1740022818000372.
Besley, Timothy, and Torsten Persson. 2014. “Why Do Developing Countries Tax So Little?” Journal of Economic Perspectives 28 (4): 99–120. http://doi.org/10.1257/jep.28.4.99.
Boeke, Julius Herman. 1953. Economics and Economic Policy of Dual Societies, as Exemplified by Polynesia. New York, NY: Institute of Pacific Relations.
Bonnet, Florence, Joann Vanek, and Martha Chen. 2019. “Women and Men in the Informal Economy: A Statistical Brief”. Manchester: Women in Informal Employment: Globalizing and Organizing.
Breusch, Trevor. 2005. “The Canadian Underground Economy: An Examination of Giles and Tedds”. Canadian Tax Journal 53 (2): 367–391.
Breusch, Trevor. 2016. “Estimating the Underground Economy Using MIMIC Models”. Journal of Tax Administration 2 (1): 41–72.
Calligaro, Florencia, and Oscar Cetrangolo. 2023. “Financing Universal Social Protection: The Relevance and Labour Market Impacts of Social Security Contributions”. WIEGO Working Paper No. 47. Manchester: Women in Informal Employment: Globalizing and Organizing.
Carroll, Erica. 2011. “Taxing Ghana’s Informal Sector: The Experience of Women”. Christian Aid Occasional Paper No. 7. London: Christian Aid.
Chen, Martha Alter. 2012. “The Informal Economy: Definitions, Theories and Policies”. WIEGO Working Paper No. 1. Manchester: Women in Informal Employment: Globalizing and Organizing.
Chen, Martha Alter, Erofili Grapsa, Ghida Ismail, Michael Rogan, Marcela Valdivia, Laura Alfers, Jenna Harvey, Ana Carolina Ogando, Sarah Orleans Reed, and Sally Roever. 2022. “COVID-19 and Informal Work: Evidence from 11 Cities”. International Labour Review 161 (1): 29–58. http://doi.org/10.1111/ilr.12221.
Deléchat, Corinne C., and Leandro Medina. 2021. The Global Informal Workforce: Priorities for Inclusive Growth. Washington, DC: International Monetary Fund.
Elgin, Ceyhun. 2021. The Informal Economy: Measures, Causes, and Consequences. 1st ed. Abingdon: Routledge.
Feige, Edgar L. 2016. “Reflections on the Meaning and Measurement of Unobserved Economies: What Do We Really Know about the ‘Shadow Economy’?” Journal of Tax Administration 2 (1): 5–40.
Frey, Bruno S., and Hannelore Weck-Hannemann. 1983. “Estimating the Shadow Economy: A ‘Naive’ Approach”. Oxford Economic Papers 35 (1): 23–44. http://doi.org/10.1093/oxfordjournals.oep.a041582.
Gallien, Max, and Vanessa van den Boogaard. 2023. “Formalization and Its Discontents: Conceptual Fallacies and Ways Forward”. Development and Change 54 (3): 490–513. http://doi.org/10.1111/dech.12768.
Gallien, Max, Giulia Mascagni, Mick Moore, Giovanni Occhiali, Wilson Prichard, Fabrizio Santoro, Celeste Scarpini, and Vanessa van den Boogaardet. 2023. “Why Mass Tax Registration Campaigns Do Not Work”. Policy Brief No. 2. Brighton: Institute of Development Studies.
Guha-Khasnobis, Basudeb, Ravi Kanbur, and Elinor Ostrom, eds. 2006. “Beyond Formality and Informality”. In Linking the Formal and Informal Economy: Concepts and Policies, edited by Basudeb Guha-Khasnobis, Ravi Kanbur and Elinor Ostrom, 1–18. Oxford: Oxford University Press.
Harris, John R., and Michael P. Todaro. 1970. “Migration, Unemployment and Development: A Two-Sector Analysis”. The American Economic Review 60 (1): 126–142.
Hart, Keith. 1973. “Informal Income Opportunities and Urban Employment in Ghana”. The Journal of Modern African Studies 11 (1): 61–89. http://doi.org/10.1017/S0022278X00008089.
Holland, Alisha C., and Calla Hummel. 2022. “Informalities: An Index Approach to Informal Work and Its Consequences”. Latin American Politics and Society 64 (2): 1–20. http://doi.org/10.1017/lap.2022.3.
Hussmanns, Ralf. 2001. “Informal Sector and Informal Employment: Elements of a Conceptual Framework”. Paper presented at ILO/WIEGO Workshop on Informal Employment Statistics in Latin America, Santiago, 16–18 October 2001.
Hussmanns, Ralf. 2005. “Measuring the Informal Economy: From Employment in the Informal Sector to Informal Employment”. ILO Working Paper No. 53. Geneva: ILO.
ILO. 1972. Employment, Incomes and Equality: A Strategy for Increasing Productive Employment in Kenya. Geneva.
ILO. 1992. Meeting of Experts on Labour Statistics, Geneva, 28 January–6 February 1992. Geneva.
ILO. 1993. Report of the Fifteenth International Conference of Labour Statisticians, Geneva, 19–28 January 1993. Geneva.
ILO. 2013. Measuring Informality: A Statistical Manual on the Informal Sector and Informal Employment. Geneva.
ILO. 2018. Women and Men in the Informal Economy: A Statistical Picture. 3rd ed. Geneva.
ILO. 2019. “Issues to be Addressed in the Revision of the Standards for Statistics on Informality”. Working Group for the Revision of the Standards for Statistics on Informality. Geneva.
ILO. 2023a. Women and Men in the Informal Economy: A Statistical Update. Geneva.
ILO. 2023b. Resolution Concerning Statistics on the Informal Economy. International Conference of Labour Statisticians. 21st Session.
ILO. 2024. “Note on the Implementation of Statistical Standards from the 19th, 20th and 21st ICLS through Household Surveys”. ILO Statistics Department Geneva.
Kirchgässner, Gebhard. 2017. “On Estimating the Size of the Shadow Economy”. German Economic Review 18 (1): 99–111. http://doi.org/10.1111/geer.12094.
Lewis, W. Arthur. 1954. “Economic Development with Unlimited Supplies of Labour”. The Manchester School 22 (2): 139–191. http://doi.org/10.1111/j.1467-9957.1954.tb00021.x.
Mascagni, Giulia, Fabrizio Santoro, Denis Mukama, John Karangwa, and Napthal Hakizimana. 2022. “Active Ghosts: Nil-Filing in Rwanda”. World Development 152 (April): Article No. 105806. http://doi.org/10.1016/j.worlddev.2021.105806.
Meagher, Kate. 2022. “Crisis Narratives and the African Paradox: African Informal Economies, COVID-19 and the Decolonization of Social Policy”. Development and Change 53 (6): 1200–1229. http://doi.org/10.1111/dech.12737.
Medina, Leandro, and Friedrich Schneider. 2021. “The Evolution of Shadow Economies through the 21st Century”. In The Global Informal Workforce: Priorities for Inclusive Growth, edited by Corinne C. Deléchat and Leandro Medina, 11–70. Washington, DC: International Monetary Fund.
Moore, Mick. 2023. “Tax Obsessions: Taxpayer Registration and the ‘Informal Sector’ in Sub-Saharan Africa”. Development Policy Review 41 (1): Article No. e12649. http://doi.org/10.1111/dpr.12649.
Neuwirth, Robert. 2011. Stealth of Nations: The Global Rise of the Informal Economy. New York, NY: Pantheon books.
Ohnsorge, Franziska, and Shu Yu, eds. 2022. The Long Shadow of Informality: Challenges and Policies. Washington, DC: World Bank Group.
Peattie, Lisa. 1987. “An Idea in Good Currency and How it Grew: The Informal Sector”. World Development 15 (7): 851–860. http://doi.org/10.1016/0305-750X(87)90038-6.
La Porta, Rafael, and Andrei Shleifer. 2008. “The Unofficial Economy and Economic Development”. NBER Working Paper No. 14520. Cambridge, MA: National Bureau of Economic Research.
Rogan, Michael. 2019. “Tax Justice and the Informal Economy: A Review of the Debates”. WIEGO Working Paper No. 41. Manchester: Women in Informal Employment: Globalizing and Organizing.
Schneider, Friedrich. 2002. “The Size and Development of the Shadow Economies of 22 Transition and 21 OECD Countries”. IZA Discussion Paper No. 514. Bonn: Institute for the Study of Labor.
Schneider, Friedrich. 2004. “The Size of the Shadow Economies of 145 Countries All Over the World: First Results over the Period 1999 to 2003”. IZA Discussion Paper No. 1431. Bonn: Institute for the Study of Labor.
Schneider, Friedrich, and Dominik H. Enste. 2000. “Shadow Economies: Size, Causes, and Consequences”. Journal of Economic Literature 38 (1): 77–114. http://doi.org/10.1257/jel.38.1.77.
de Soto, Hernando. 1989. The Other Path: The Economic Answer to Terrorism. New York, NY: Basic Books.
Teal, Francis. 2023. “Firm Size, Employment and Value Added in African Manufacturing Firms: Why Ghana Needs Its 1%”. Journal of African Economies 32 (2): 118–136. http://doi.org/10.1093/jae/ejab015.
Ulyssea, Gabriel. 2020. “Informality: Causes and Consequences for Development”. Annual Review of Economics 12 (1): 525–546. http://doi.org/10.1146/annurev-economics-082119-121914.
WIEGO (Women in Informal Employment: Globalizing and Organizing). 2020. “Rethinking Formalization”. WIEGO Blog. https://www.wiego.org/rethinking-formalization.
Williams, Colin C. 2015. “Out of the Shadows: Classifying Economies by the Extent and Nature of Employment in the Informal Economy”. International Labour Review 154 (3): 331–351. http://doi.org/10.1111/j.1564-913X.2015.00245.x.
Williams, Colin C. 2023. A Modern Guide to the Informal Economy. Cheltenham: Edward Elgar Publishing Limited.