shadow economies: size, causes, and · pdf fileschneider and enste: shadow economies: size,...

Journal of Economic Literature Vol. XXXVIII (March 2000) pp. 77–114

Shadow Economies: Size, Causes, and Consequences

FRIEDRICH SCHNEIDER and DOMINIK H. ENSTE1

1. Introduction

SHADOW OR UNDERGROUND eco-nomic activity is a fact of life around

the world, and there are strong indica-tions that it is increasing. Most societiesattempt to control these activitiesthrough various punitive measures orthrough education, rather than throughreforms of the tax and social security sys-tems which could improve the dynamicsof the official economy. Gathering infor-mation about underground economic ac-tivity is difficult, because no one en-gaged in such activity wants to beidentified. Obtaining accurate statisticsabout the allocation of a country’s re-sources in the shadow economy is impor-tant for making effective economic pol-icy decisions. Hence, it is crucial to knowwho is engaged in the shadow economy,and with what frequency and magnitudesuch activities occur.

Although a large literature exists onsingle aspects of the hidden economy, acurrent comprehensive survey is lack-ing. Disagreement persists about defini-tions and estimation procedures, andabout the use of estimates in economicanalysis and policy. The feature “Con-troversy: On the Hidden Economy” inEconomic Journal (Vol. 109, No. 456,June 1999) documents the differingopinions of, e.g., Vito Tanzi (1999),James J. Thomas (1999), and David E.A. Giles (1999a). The size, causes, andconsequences of the shadow economyvary for different types of countries, butsome comparisons can be made whichmight be useful for social scientists andpoliticians, who must deal with thisphenomenon sooner or later.

There are several important reasonswhy policy makers should be especiallyconcerned about the rise of the shadoweconomy. Among the most important ofthese are:

(i) A growing shadow economy can beseen as the reaction of individualswho feel overburdened by the stateand who choose the “exit option”rather than the “voice option” (Al-bert O. Hirschman 1970). If the in-crease of the shadow economy iscaused by a rise in the overall tax andsocial security burden together with“institutional sclerosis” (Mancur

77

1 Schneider: Johannes Kepler University ofLinz, Austria. Enste: University of Cologne, Ger-many. A first version of this paper was written dur-ing a stay of the first author at the IMF, Washing-ton, D.C., and of the second author at the Centerfor Study of Public Choice, George Mason Univer-sity, Fairfax, Virginia. We are grateful to RichardHemming (IMF), Hendrik Juerges (University ofDortmund, Germany), Daniel Kaufmann (WorldBank), Norman Loayza (World Bank), MelanieMuellenmeister (University of Cologne), ParthoShome (IMF), Vito Tanzi (IMF), Pablo Zoido-Lobatón (World Bank) and two anonymous refe-rees of this journal for their helpful comments andstimulating criticism of earlier drafts.

Olson 1982), then the “consecutiveflight” into the shadow economymay erode the tax and social secu-rity bases. The result can be a vi-cious circle of a further increase inthe budget deficit or tax rates, addi-tional growth of the shadow econ-omy, and gradual weakening ofthe economic and social basis ofcollective arrangements.2

(ii) A prospering shadow economy maycause severe difficulties for politi-cians because official indicators—onunemployment, labor force, in-come, consumption—are unreli-able. Policy based on erroneous of-ficial indicators is likely to beineffective, or worse.

(iii) The effects of a growing shadoweconomy on the official one mustalso be considered. On the onehand, a prospering shadow economymay attract (domestic and foreign)workers away from the official econ-omy and create competition for offi-cial firms. On the other hand, atleast two-thirds of the incomeearned in the shadow economy isimmediately spent in the officialeconomy, thus having a positiveeffect on the official economy.3

We undertook the task of collectingavailable data on the shadow economyto determine its development and sizeover an extended period of time for as

many countries as possible. In Section 2,we attempt to define shadow economyactivities and present an overview ofsome empirical results. In Section 3 weexamine the main causes of the develop-ment of the shadow economy, and inSection 4 we analyze the interactions be-tween the official and unofficial econo-mies. Section 5 provides a preliminaryanalysis of the link between corruptionand the shadow economy, and in Section6 the various methods of estimating thesize of the shadow economy are pre-sented. In Section 7 we discuss more de-tailed empirical findings on the size ofthe shadow economy for developing,transition, and OECD countries. Finally,in Section 8 we summarize and drawsome conclusions.

2. The Shadow Economy: Definition and Size

2.1 What Is the Shadow Economy?

Attempts to measure the shadoweconomy first face the problem of de-fining it. One commonly used workingdefinition is: all economic activities thatcontribute to the officially calculated(or observed) gross national product butare currently unregistered. This defini-tion is used, for example, by Edgar L.Feige (1989, 1994), Schneider (1994a),Frey and Werner Pommerehne (1984),and Herald Lubell (1991). Philip Smith(1994, p. 18) defines it as “market-based production of goods and services,whether legal or illegal, that escapes de-tection in the official estimates ofGDP.” But these definitions fall shortof addressing all questions. Table 1 mightbe helpful in developing a reasonableconsensus definition of the legal andillegal shadow economy.

According to Table 1, the shadoweconomy includes unreported incomefrom the production of legal goods andservices, either from monetary or barter

2 For further analysis of the impacts of theshadow economy, see Schneider and Enste (2000).Some more general implications for governmentsare discussed e.g. by Bruno S. Frey (1997); Frey,Felix Oberholzer-Gee, and Reiner Eichenberger(1996); Frey and Eichenberger (1996); and ElinorOstrom (1990).

3 This figure was derived from polls of the Ger-man and Austrian populations about the (effectsof) the shadow economy. For further informationsee Schneider (1998b). These polls also show thattwo-thirds of the value added produced in theshadow economy would not be produced in the offi-cial economy without the activities in the shadoweconomy.

78 Journal of Economic Literature, Vol. XXXVIII (March 2000)

transactions, hence all economic activi-ties that would generally be taxablewere they reported to the tax authori-ties. In general, a precise definitionseems quite difficult, if not impossible,as “the shadow economy develops allthe time according to the ‘principle ofrunning water’: it adjusts to changes intaxes, to sanctions from the tax authori-ties and to general moral attitudes, etc.”(Gunnar Mogensen, Hans K. Kvist,Eszter Körmendi, and Soren Pedersen1995, p. 5).4

Moreover, the definition often variesdepending on the chosen method of mea-surement. In our analysis we concentrateon legal value-added creating activitieswhich are not taxed or registered andwhere the largest part can be classifiedas “black” or clandestine labor.5

Our survey does not focus on tax

evasion itself, but rather serves as a sup-plement to the recent survey on taxcompliance by James Andreoni, BrianErard, and Jonathan S. Feinstein (1998,p. 819) which explicitly excludes theshadow economy: “Unfortunately, thereare many important issues that we donot have room to discuss, most notablythe vast literature on the undergroundeconomy which exists in part as a meansof evading taxes.”6 Still, there are someconnections between these two researchareas. See for example Feinstein (1999),who tries to close the gap between taxevasion and shadow economy research.

TABLE 1A TAXONOMY OF UNDERGROUND ECONOMIC ACTIVITIES

Monetary Transactions Nonmonetary Transactions

Illegal Activities Trade in stolen goods; drug dealing andmanufacturing; prostitution; gambling;smuggling and fraud.

Barter: drugs, stolen goods, smuggling, etc.Produce or growing drugs for own use.Theft for own use.

Tax Evasion Tax Avoidance Tax Evasion Tax Avoidance

Legal Activities Unreported incomefrom self-employment;Wages, salaries andassets fromunreported workrelated to legalservices and goods

Employeediscounts, fringebenefits

Barter of legalservices and goods

All do-it-yourselfwork and neighborhelp

Source: Rolf Mirus and Roger S. Smith (1997, p. 5), with additional remarks.

4 For a detailed discussion, see “Controversy:On the Hidden Economy” in Economic Journal(Vol. 109, no. 456, June 1999), Frey and Pom-merehne (1984); Feige (1989); Thomas (1992,1999); and Schneider (1986, 1994a, 1998a).

5 This means that unpaid or “pure” householdproduction, voluntary nonprofit (social) servicesand criminal activities are excluded from theanalysis. See Thomas (1992) for a broader view

and comprehensive analysis of the household, in-formal, irregular, and criminal sectors in differenttypes of countries.

6 While there have been many theoretical stud-ies on tax evasion in the last twenty years, empiri-cal studies are harder to find. Many are based ontax compliance experiments and cover only partsof the shadow economy. Convincing empirical evi-dence for the theoretical hypothesis on why peo-ple evade taxes is hard to find, and the empiricalresults are ambiguous (Pommerehne and Han-nelore Weck-Hanneman 1992). James Alm (1996)gives an overview of tax compliance explanationsin different studies. The theoretical literature ontax evasion is summarized in Frank Cowell (1990);see also Michael G. Allingham and Agnar Sandmo(1972) for their pathbreaking study in this area.

Schneider and Enste: Shadow Economies: Size, Causes, and Consequences 79

2.2 How Large Is the Shadow Economy?

A main focus of this survey is to givea comprehensive summary of availabledata on the size of the shadow economy,since there has been no consistent com-parison of estimates on various coun-tries generated using similar methods.An overview of some results, estimatedwith indirect or “indicator” methods, isgiven in Tables 2 and 3, which provideapproximate magnitudes of the size anddevelopment of the underground econ-omy, defined as productive value-addingactivities that should be included in theofficial GNP.7

Table 2 provides a rough comparisonof the size of the shadow economy rela-tive to official GDP for a selection ofdeveloping, transition, and OECDeconomies in the early 1990s, using thephysical input (electricity) and currency

7 The more detailed results for seventy-six de-veloping, transition, and OECD countries can be

found in Section 7 (including the sources). Thedifferent methods used to measure the size of theshadow economy are described in Section 6.


demand approaches. The estimates forsome of these countries (Nigeria,Egypt, and Thailand) show an under-ground sector that is nearly three-quar-ters the size of officially recorded GDP.In many countries, especially in Centraland South America, the size is one-quarter to one-third of GNP. In Asiancountries, with a comparatively smallpublic sector, high tax morale and/orhigh expected punishment (for examplein Hong Kong and Singapore), theshadow economy is estimated to besimilar to that in many “northern”European countries. Transition econo-mies are estimated to often have sub-stantial unofficial activity, many ataround one-quarter of GNP. The big-gest shadow economies belong to someof the former Soviet Union transitioncountries (between 28–43 percent ofGDP), like Georgia, Ukraine and Bela-rus. The Czech Republic can be foundat the lower end; according to theseestimates, the underground sector isaround 10 percent of GDP.

Turning to the OECD countries inSouthern Europe, Greece and Italyhave underground economies almostone-third as large as officially measuredGNP. Spain, Portugal, and Belgiumhave shadow economies between 20–24

percent of (official) GNP. According tothese estimates, the Scandinavian coun-tries also have sizeable unofficial econo-mies (between 18–20 percent of GNP)which is attributed mainly to the highfiscal burden. The “central” Europeancountries (Ireland, the Netherlands,France, Germany and Great Britain)have smaller shadow economies (be-tween 13–16 percent of GNP), probablydue to lower fiscal burdens and moder-ate regulatory restrictions. The smallestunderground economies are estimatedto exist in countries with relativelysmall public sectors (Japan, the US, andSwitzerland) and comparatively high taxmorale (the US and Switzerland).

Table 3 reports estimates of thegrowth of the underground economy(relative to GNP) for selected Westerncountries and the US, using the cur-rency demand approach. The Scandina-vian countries (Sweden, Norway, andDenmark) and the German-speakingcountries (Germany and Austria) exhibita sizeable increase of their undergroundeconomies within the thirty-five yearscovered (1960–95). The countries witha low share (Switzerland, Austria, andthe US) also show a significant increase;in all three countries the share morethan doubled. Sizeable increases have

TABLE 3GROWTH OF SHADOW ECONOMY RELATIVE TO GNP

SELECTED WEST EUROPEAN COUNTRIES AND THE UNITED STATES, 1960–95

Size of Shadow EconomyIncrease in Shadow

Country 1960 1995 Economy

Sweden 2% 16% 14%Denmark 4.5% 17.5% 13%Norway 1.5% 18% 16.5%Germany 2% 13.2% 11.2%United States 3.5% 9.5% 6%Austria 0.5% 7% 6.5%Switzerland 1% 6.7% 5.7%

Source: Authors’ calculations based on the currency demand approach (rounded figures).


been estimated, with few exceptions, forall types of countries and with all kindsof approaches: the increasing impor-tance of the underground relative to theofficial economy is a robust phenomenon(see Section 7).

3. The Main Causes of the Increase

The growth of the shadow economy iscaused by many different factors. Themost important and often cited onesare:8 the rise of the burden of taxes andsocial security contributions; increasedregulation in the official economy, espe-cially of labor markets; forced reductionof weekly working time; earlier retire-ment; unemployment; and the decline ofcivic virtue and loyalty towards publicinstitutions combined with a decliningtax morale.

An interdisciplinary analysis of thecauses of the increase of the shadoweconomy seems necessary, since eco-nomic factors can only partly explainthe increase.9 Micro-sociological andpsychological approaches can provideinteresting additional insights in thedecision making process of individ-uals choosing to work underground(Schneider and Enste 2000). In an in-terdisciplinary approach (as undertakenin economic psychology), variables suchas tax morale (first discussed by GünterSchmölders 1960, 1975) and acceptanceand perceived fairness of the tax system

are considered. A discussion of the im-portance of interdisciplinary researchcan be found in recent articles in thisjournal by Matthew Rabin (1998), JonElster (1998), and Shira B. Lewin(1996). For a broader view see RobertH. Frank (1988) and Frey (1997). How-ever, since our article concentrates oneconomic factors, we will focus on theeconomic reasoning.

3.1 The Burden of Tax and Social Security Contributions

In almost all studies, one of the mostimportant causes of the increase of theshadow economy is the rise of the taxand social security burdens.10 Sincetaxes affect labor–leisure choices, andalso stimulate labor supply in theshadow economy (the untaxed sector ofthe economy), the distortion of thischoice is a major concern of econo-mists. The bigger the difference be-tween the total cost of labor in the offi-cial economy and after-tax earnings(from work), the greater the incentiveto avoid this difference and to work inthe shadow economy. Since this differ-ence depends broadly on the social se-curity system and the overall tax bur-den, they are key features of the existenceand rise of the shadow economy.

A recent macroeconomic analysis ofthe matter is given by Norman V.Loayza (1996). He presents a simplemacroeconomic endogenous growthmodel whose production technology de-pends on congestable public services.The determinants and effects of exces-sive taxes and regulations on the infor-mal sector are studied, where the gov-ernment lacks the capability to enforce

8 When dealing with the various causes in sec-tions 3.1 to 3.5, the most important references aregiven. For an overall view, see the studies byDieter Cassel and E. Ulrich Cichy (1986); Tanzi(1982); Frey and Pommerehne (1984); Thomas(1992), and Schneider and Enste (2000).

9 Although thus far interdisciplinary research fo-cuses on tax compliance (see Alm, Gary McClel-land, and William Schulze 1999; Cowell 1990;Pommerehne, Albert Hart, and Frey 1994; and thespecial issue on “Economic Psychological Perspec-tives on Taxation” of the Journal of Economic Psy-chology, Dec. 1992), interdisciplinary approachescan also be used to explain other hidden activities;see Frey (1997).

10 See Tanzi (1982, 1999); Frey and Pomme-rehne (1984); Feige (1989); Susan Pozo (1996);Owen Lippert and Michael Walker (1997);Schneider (1994a,b, 1997, 1998a, 1999); Schneiderand Enste (2000); Thomas (1992); Hernando DeSoto (1989); Ben-Zion Zilberfarb (1986); and Giles(1999a) .


compliance. Using the Multiple-Indica-tor-Multiple-Cause (MIMIC) model(see Section 6.3), his empirical ap-proach treats the informal sector as anunobserved variable for which multiplecauses and indicators exist. He esti-mates the size of the informal sector infourteen Latin American countries andfinds some evidence for three determi-nants being significantly relevant at the10 percent confidence level. Tax bur-den (0.33) and labor market restrictions(0.49) affect the relative size of the in-formal sector positively, while thestrength and efficiency (–0.42) of gov-ernment institutions have a negativeinfluence leading to a decrease of theinformal sector.11 Because Loayza’s ap-proach only shows statistical correla-tions rather than causal relations, hecan only partly answer questions like:Why do people choose to work in theshadow economy? What other factors(besides income motive) cause an in-crease of informal activities? Can othertheories provide further help in deter-mining relevant factors? Since, accord-ing to methodological individualism,only individuals can choose, it might behelpful to take a closer look at the indi-vidual decision (with respect to the in-fluence of the tax and social securityburden) to work in the shadow economy.

The determinants for a household towork in the shadow economy are similarto those of tax evasion, namely: howmuch income should be declared to thetax authorities. Reinhard Neck, MarkusHofreither, and Schneider (1989) inves-tigated the determinants of a house-hold’s supply of underground labor andits demand for underground goods,showing that under an additive-separa-ble utility function and with a two-stage

decision of the consumer, higher mar-ginal income tax rates imply a highersupply of underground labor, andhigher wage rates in the official econ-omy imply a lower supply of under-ground labor. On the other hand, theyshowed that the firms’ demand for un-derground labor and supply of under-ground goods depend positively on theindirect tax and wage rates in the offi-cial economy (under the assumption offixed nonhuman factors of production,and separate production functions forofficial and underground goods). Disre-garding other factors influencing the ex-tent of the shadow economy, one canconjecture that higher indirect tax ratesand higher marginal income tax ratestend to raise the amount of labor andgoods bought and sold in the under-ground sector. Official sector wage ratechanges may have a positive or negativeinfluence on the equilibrium amount ofunderground labor, depending uponwhether demand or supply changesdominate. In addition, the equilibriumquantities of shadow economy labor andgoods also depend on other variables,like penalty rates and detection prob-abilities for tax evasion, which are tosome extent under government control.

One must, however, be very carefulnot to draw premature policy conclu-sions from such a model. First, the com-parative static results do not generalizeto arbitrary utility and production func-tions. Second, the analysis concentratedon the determinants of the quantities ofgoods and labor supplied and demandedby individual firms and households, anddid not analyze market equilibrium con-ditions. The model ought to be closedby putting individual decision makersinto the context of a general equilib-rium model, with at least two labor mar-kets and two goods markets, the officialand the shadow economy markets ineach case. Only in such a framework

11 The numbers indicate the change of the sizeof the informal sector (in standard deviations) witha one standard deviation increase in each of thedeterminants.


could all spillovers be analyzed appro-priately, and prices and wages, whichare assumed to be given for the individ-ual transactors, could be determinedendogenously. Also, on a more generallevel, an analysis of the effects of thedegree of progression on the shadoweconomy has to take into account differ-ences of reactions across consumerswith respect to the total and official la-bor supply. The different effects on theofficial and underground labor supplyso far seem to be an open question,which could be appropriately treated ina general equilibrium model, with offi-cial and underground markets for laborand goods and with different types ofconsumers. To our knowledge, such atheoretical model is not available at thepresent time.

In another study, Schneider andNeck (1993) investigate how the com-plexity of the tax system affects the sizeof the shadow economy. A complex in-come tax schedule allows more legal taxavoidance than a simple one by provid-ing various tax exemptions and reduc-tions. According to this view, a compre-hensive income tax displays a lowdegree of complexity. Schneider andNeck show in their theoretical modelthat a more complex tax system implies,ceteris paribus, a smaller labor supplyin the shadow economy, because a morecomplex tax system makes individual ef-forts to legally avoid taxation more prof-itable. At the same time, it encourageshouseholds to work in the official econ-omy instead of the underground, be-cause the reduced tax burden makes taxevasion (with the risk of being caughtand punished) less attractive. Broaden-ing the income tax base and removingtax exemptions can therefore increasethe size of the shadow economy. TheAustrian tax reform of 1989 not only re-duced marginal income tax rates butalso broadened the tax base by abolish-

ing several exemptions and loopholes,producing a less complex tax system.Schneider and Neck empirically analyzethe effects of changing tax structureson the development of the Austrianshadow economy. One would expectthat a decrease in direct taxes wouldlead to a decline in the shadow econ-omy; such a result was actually notfound. The explanation offered bySchneider and Neck was that not onlyare direct and indirect taxes an impor-tant factor influencing the shadoweconomy, but the complexity of the taxsystem and the burden of regulation areimportant as well. The theoretical andempirical results in their study indicatethat both factors—i.e. a less complextax system with a broader tax base, andincreased regulation—more than offsetthe lower tax burden in 1989.12

The influence of indirect and directtaxation on the shadow economy canbe further demonstrated by discussingempirical results on Austria and theScandinavian countries. In the case ofAustria, Schneider (1994b) estimates acurrency demand function including asdriving forces for the shadow economythe following four variables: direct taxa-tion; indirect taxation; complexity of thetax system; and intensity of governmentregulations. The direct tax burden (in-cluding social security payments) hasthe biggest estimated influence, fol-lowed by the intensity of regulation andcomplexity of the tax system on the cur-rency demand. A similar result was ob-tained by Schneider (1986) for Den-mark, Norway, and Sweden. In all threecountries, tax variables (average directtax rate, average total tax rate [indirect

12 For Canada, Peter S. Spiro (1993) finds thatpeople once working in the shadow economy likethe high profiles from irregular activities, developsocial networks and personal relationships andhence will not return to the official economy evenin the long run.


and direct] and marginal tax rates) havethe expected positive influence (on cur-rency demand) and are statistically sig-nificant. Similar results were reachedby Gebhard Kirchgaessner (1983, 1984)for Germany and by Jan Klovland(1984) for Norway and Sweden.

Two other recent studies provide evi-dence of the influence of income taxeson the shadow economy. Richard J.Cebula (1997), using Feige’s data, foundsome impact of income tax rates, IRSaudit probabilities, and penalty policieson the relative size of the shadow econ-omy in the United States. He concludesthat restraining from increasing the topmarginal income tax rate may preventfurther increase of the shadow economy,while increased IRS audits and penal-ties might reduce the shadow economy;his findings indicate that governmentactions generally have a strong influ-ence. For example, if the marginal fed-eral personal income tax rate increasesby one percentage point, ceteris pari-bus, the shadow economy rises by 1.4percentage points. In another investiga-tion, Roderick Hill and Muhammed Kabir(1996) found empirical evidence thatmarginal tax rates are more relevant thanaverage tax rates, and that a substitutionof direct taxes by indirect taxes seemsunlikely to improve tax compliance.

More evidence on the effect of taxa-tion on the shadow economy is pre-sented by Simon Johnson, Daniel Kauf-mann, and Pablo Zoido-Lobatón(1998a,b), who conclude that it is nothigher tax rates per se that increase thesize of the shadow economy, but inef-fective and discretionary application ofthe tax system and regulations by gov-ernment. Their finding, that there is anegative correlation between the size ofthe unofficial economy and the top (mar-ginal) tax rates, might be unexpected,but since other factors—like tax deduc-tibility, tax relief, tax exemptions, the

choice between different tax systems,and various other options for legal taxavoidance—were not taken into ac-count, it is not all that surprising. Simi-larly, Eric Friedman, Johnson, Kaufmannand Zoido-Lobatón (1999) found in across-country analysis that higher taxrates are associated with less unofficialactivity as percent of GDP. They arguethat entrepreneurs go underground notto avoid official taxes but to reduce theburden of bureaucracy and corruption.However, looking at their empirical (re-gression) results, the finding that highertax rates are correlated with a lowershare of the unofficial economy is notvery robust, and in most cases, usingdifferent tax rates, they do not find astatistically significant result. The over-all conclusion of the studies is thatthere is a large difference between theimpact of either the direct tax or thecorporate tax burden, and institutionalaspects like efficiency of the adminis-tration, the extent of control by politi-cians and bureaucrats, the amount ofbribery, and especially corruption.Johnson, Kaufmann, and Zoido-Lobatón(1998b) argue that these aspects play abigger role in the bargaining game be-tween the government and the taxpayersthan the tax burden.

3.2 Intensity of Regulation

Increased regulation reduces indi-viduals’ choices in the official econ-omy.13 Intensity of regulation is oftenmeasured by the number of laws and re-quirements such as licenses, and onecan think of labor market regulations,labor restrictions for foreigners, andtrade barriers. The influence of labor

13 For a psychological foundation of this feature(theory of reactance), see Jack W. Brehm (1966,1972), and for a (first) application to the shadoweconomy Linde Pelzmann (1988). See Schneiderand Enste (2000) for an integration of this theoryin an interdisciplinary (rational choice) approach.


regulations on the shadow economy inGermany is described in DeregulationCommission (1991) and Monopol-kom-mission (1998). Regulations lead to in-creased labor costs in the official econ-omy. Since most of these costs can beshifted onto employees, they provideanother incentive to work in the shadoweconomy, where they can be avoided.Schneider and Günther Pöll (1999) pre-sent some empirical evidence of thisimpact.

The model of Johnson, Kaufmann,and Andrei Shleifer (1997) predicts, in-ter alia, that countries with more gen-eral regulation of their economies tendto have a higher share of the unofficialeconomy in total GDP. A one-point in-crease of the regulation index (rangingfrom 1 to 5, with 5 = the most regula-tion in a country), ceteris paribus, is as-sociated with an 8.1 percentage pointincrease in the share of the shadoweconomy, when controlled for GDP percapita (Johnson, Kaufmann, and Zoido-Lobatón 1998b, p. 18). They concludethat the enforcement of regulation—rather than the overall extent of regula-tion (mostly not enforced)—is the keyfactor for the burden levied on firmsand individuals that drives them intothe shadow economy. Friedman,Johnson, Kaufmann, and Zoido-Lobatón(1999) reach a similar result. In theirstudy, every available measure of regu-lation is significantly correlated withthe share of the unofficial economy, andthe direction of the correlation is unam-biguous: more regulation is correlatedwith a larger shadow economy. A one-point increase in an index of regulation(ranging from 1–5) is associated witha 10–percent increase in the shadoweconomy for seventy-six developing,transition, and developed countries.14

These findings demonstrate that gov-ernments should put more emphasis onreducing the density of regulations or atleast on improving enforcement of lawsand regulations, instead of increasingthe number of regulations. Some gov-ernments, however, opt for more regu-lation and laws in trying to reduce theshadow economy, mostly because itleads to increased power for bureau-crats and to higher employment in thepublic sector. Some politicians may nothave a sincere interest in substantiallyreducing the shadow economy, sincemany voters gain from unofficial activi-ties. The signaling of “fighting for lawand order” might therefore be moreuseful for getting politicians reelectedthan would deep reforms of the tax andsocial security systems.15

3.3 Social Transfers

The social welfare system leads tostrong negative incentives for benefici-aries to work in the official economy,since their marginal tax rate often ap-proaches or equals 100 percent. Thiscan be derived from the neoclassical lei-sure-income model presented by Peterde Gijsel (1984); Volker Riebel (1983,1984); and Schneider and Enste (2000).For Canada, see Thomas Lemieux, Ber-nard Fortin, and Pierre Fréchette(1994). For Germany, Siegfried Lam-nek, Gaby Olbrich, and Wolfgang Schäfer(1999) found empirical evidence of thisimpact. Such a system provides disin-centives for individuals receiving wel-fare payments to even search for workin the official economy, since theiroverall income is higher if they receivethese transfers while working in theunderground economy.

14 De Soto (1989) in his famous book describesin more detail the costs of regulation in Peru.

15 See Frey (1989) for a first application of thepublic choice theory to the shadow economy, andfor further discussion Cassel (1989), andSchneider and Enste (2000).


3.4 Labor Market

Overregulation and labor costs in theofficial labor market are driving forcesfor the shadow economy. Two main as-pects—the reduction of official workinghours, and the unemployment rate—arediscussed quite often in this context. Inmost OECD countries, unemploymentis to a large extent caused by high totallabor costs. This can also be seen as acause for the increase of the shadoweconomy.

The reduction in working hours inthe official economy was introduced bygovernments (e.g. France) and/or laborunions (e.g. Germany) in order to re-duce unemployment. An overview ofthese economic policy measures is givenin OECD (1998, pp. 123–88). Thethinking behind the policy is that thereis a limited quantity of work, which hasto be redistributed. But this idea ne-glects a key factor: a forced reductionof working hours contrary to employeepreferences increases the potentialhours they can work in the shadoweconomy (see for example JenniferHunt 1999).16 Early retirement andpart-time work also offer opportunitiesfor individuals to work in the untaxed,unregulated economy (de Gijsel 1984;Riebel 1983, 1984). The redistributionof work can be successful only if it is inaccordance with individual preferencesfor leisure or if individuals are incapa-ble of working. Otherwise, they maychoose to work more—underground.17

More detailed information on the la-bor supply decision in the underground

economy is given by Lemieux, Fortin,and Fréchette (1994) using micro datafrom a survey conducted in QuebecCity (Canada). The results of theirstudy suggest that hours worked in theshadow economy are responsive tochanges in the net wage in the regularsector. Their empirical findings indicatethat “participation rates and hoursworked in the underground sector alsotend to be inversely related to the num-ber of hours worked in the regular sec-tor” (Lemieux, Fortin, and Fréchette1994, p. 235). Their results emphasizea large negative elasticity of hoursworked in the shadow economy with re-spect to the wage rate in the regularsector, and also a high mobility betweenthe sectors. A (further) reduction of(official) working hours can thereforelead to an increase in the shadow econ-omy, since—for example in Germany—almost all recent empirical investigationsshow that most employees do not wantfurther reduction (Schneider and Enste2000; DIW 1998; Bosch and Lehndorff1998). Hence, a reasonable economicpolicy suggestion is more flexible work-ing hours in accordance with employeepreferences, because this would reducedistortion of the individual decision.

3.5 Public Sector Services

An increase in the shadow economyleads to decreased state revenue, whichin turn reduces the quality and quantityof publicly provided goods and services.Ultimately, this can lead to increasedtax rates in the official sector, oftencombined with deterioration in thequality of public goods (such as thepublic infrastructure) and of the ad-ministration, with the consequence ofeven stronger incentives to participatein the shadow economy. Johnson, Kauf-mann, and Zoido-Lobatón (1998b) pre-sent a simple model of this relationship.Their findings show that smaller shadow

16 After Volkswagen in Germany reduced work-ing hours considerably, there was some (thus farbasically anecdotal) evidence that in the areaaround the firm, much more reconstruction andrenovation of houses took place than in othersimilar regions.

17 See Gary S. Becker (1965) for the theoreticalfoundation, and F. Thomas Juster and Frank P.Stafford (1991) for a more detailed analysis of theallocation of time.


economies appear in countries withhigher tax revenues, if achieved bylower tax rates, fewer laws and regula-tions, and less bribery facing enter-prises. Countries with a better rule oflaw which is financed by tax revenuesalso have smaller shadow economies.Transition countries have higher levelsof regulation leading to a significantlyhigher incidence of bribery, higher ef-fective taxes on official activities, and alarge discretionary framework of regula-tions—and consequently, larger shadoweconomies. Their overall conclusion isthat “wealthier countries of the OECD,as well as some in Eastern Europe, findthemselves in the ‘good equilibrium’ ofrelatively low tax and regulatory bur-dens, sizeable revenue mobilization, goodrule of law and corruption control, and(relatively) small unofficial economy. Bycontrast, a number of countries of LatinAmerica and the former Soviet Unionexhibit characteristics consistent with a‘bad equilibrium’: tax and regulatorydiscretion and burden on the firm arehigh, the rule of law is weak, and thereis a high incidence of bribery and arelatively high share of activities in theunofficial economy” (Johnson, Kaufmannand Zoido-Lobatón 1998a).

In many countries, therefore, thepublic sector faces the challenge of sub-stantially reforming the social securityand tax systems to prevent the total de-feat of the protective welfare state be-cause of the vicious circle: high tax andregulatory burdens cause an increase inthe shadow economy, bringing addi-tional pressure on public finance, re-sulting in higher tax rates, which againincrease the incentives to evade taxesand to escape into the shadow economy,and so on. In a cumulative process, ex-isting institutions and rules might loseacceptance in the society, resulting ina situation where democratic voting(voice) is less attractive than using the

exit option—the shadow economy.Eventually, loyalty to democratic politi-cal institutions is abandoned or cannotdevelop, as can be seen in some formerSoviet Union states. The shadow econ-omy can therefore be seen as a “chal-lenge to the welfare state” (Manfred E.Streit 1984).

4. Effects of the Shadow Economy on the Official Economy

In order to study the effects of theshadow economy on the allocation of re-sources, several studies integrate under-ground economies into macroeconomicmodels.18 John F. Houston (1987) de-velops a theoretical model of the busi-ness cycle as well as tax and monetarypolicy linkages with the shadow econ-omy. He concludes that, on the onehand, the shadow economy’s effectshould be taken into account in settingtax and regulatory policies, and on theother hand, the existence of a shadoweconomy could lead to overstatement ofthe inflationary effects of fiscal ormonetary stimuli. In their study for Bel-gium, Markus C. Adam and VictorGinsburgh (1985) focus on the implica-tions of the shadow economy on officialgrowth. They find a positive relation-ship between the growth of the shadoweconomy and the official one, and theyconclude under certain assumptions(i.e. low entry costs into the shadoweconomy due to low probability of en-forcement) that an expansionary fiscalpolicy is a positive stimulus for both theformal and informal economies. A studyfor the United States by Ronald Fich-tenbaum (1989) argues that the US pro-ductivity slowdown over the period1970–89 was overstated, as underre-porting of income due to the more

18 For Austria this was done by Schneider, Hof-reither, and Neck (1989) and Neck, Hofreither,and Schneider (1989). For further discussion ofthis see Peter J. Quirk (1996) and Giles (1999a).


rapid growth of the US shadow econ-omy during this period was not takeninto account. Similar impacts werefound by Pommerehne and Schneider(1985).

Another hypothesis is that a substan-tial reduction of the shadow economyleads to a significant increase in taxrevenues and therefore to a greaterquantity and quality of public goods andservices, which ultimately can stimulateeconomic growth. Some authors foundevidence for this hypothesis. Loayza(1996) concludes that in economieswhere (1) the statutory tax burden islarger than optimal, and where (2) en-forcement of compliance is too weak,the increase in the relative size of theinformal economy generates a reductionof economic growth. The reason for thisis the negative correlation between theinformal sector and public infrastruc-ture indices, while public infrastructureis the key element for economic growth.For example, Loayza finds empiricalevidence for Latin American countriesthat if the shadow economy increasesby one percentage point (of GDP)—ce-teris paribus—the growth rate of offi-cial real GDP per capita decreases by1.2 percentage points.

This negative impact of informal sec-tor activities on economic growth is notbroadly accepted. For example, the keyfeature of the model has been criti-cized. The model is based on the as-sumption that the production technol-ogy depends on tax-financed publicservices which are subject to conges-tion; that is contrary to the general defi-nition of public goods, which are notsubject to congestion (unlike privategoods). In addition, the informal sectordoes not pay taxes but must pay penal-ties which are not used to finance pub-lic services. The negative correlationbetween the size of the informal sectorand economic growth is therefore not

surprising. Patrick K. Asea (1996) givesa more detailed criticism of the Loayzamodel.

Depending on the prevailing view ofthe informal sector, one might alsocome to the opposite conclusion. In theneoclassical view, the undergroundeconomy, responding to the economicenvironment’s demand for urban ser-vices and small-scale manufacturing,adds to the economy a dynamic and en-trepreneurial spirit and can lead tomore competition, higher efficiency,and limits on government activities. Theinformal sector may also contribute “tothe creation of markets, increase finan-cial resources, enhance entrepreneur-ship, and transform the legal, social,and economic institutions necessary foraccumulation” (Asea 1996, p. 166). Thevoluntary self-selection between theformal and informal sectors may pro-vide a higher potential for economicgrowth and, hence, a positive correla-tion between an increase in the informalsector and economic growth. The effectsof the shadow economy on economicgrowth therefore remain ambiguous.

The empirical evidence on these op-posite hypotheses is also not clear.Since many Latin American countrieshad or still have excessive regulationand weak government institutions,Loayza (1996) finds evidence for theimplications of his growth model in theearly 1990s in these countries. An in-crease in the size of the shadow econ-omy negatively affects growth (1) by re-ducing the availability of public servicesfor everyone, and (2) by using the exist-ing public services less efficiently or notat all. But the positive side effects ofshadow economy activities must be con-sidered, too. Empirical findings bySchneider (1998b) show that over 66percent of earnings in the shadow econ-omy are immediately spent in the offi-cial sector, with positive effects for


economic growth and for indirect taxrevenues. Dilip K. Bhattacharyya (1993,1999) found evidence for the UnitedKingdom (1960–84) that the hiddeneconomy has a positive effect on con-sumer expenditures of nondurable goodsand services, and an even stronger posi-tive effect on consumer expenditures ondurable goods and services. A close in-teraction between official and unofficialeconomies is also emphasized in Giles(1999a) and Tanzi (1999).

5. Corruption and the ShadowEconomy—Substitutive orComplementary Effects?

Over the last ten years, corruptionhas gained growing attention among sci-entists, politicians, and public officials.Its origins and consequences and waysto fight it have been analyzed. The lit-erature is quite large and only some (re-cent) publications can be mentioned here:Susan Rose-Ackermann (1978, 1997,1999); Arvind Jain (1998); Shleifer andRobert W. Vishny (1993); Tanzi (1994,1998); Tanzi and Hamid Davoodi (1997);Johnson, Kaufmann, and Zoido-Lobatón(1998a,b); Kaufmann and Jeffrey Sachs(1998); for the latest survey see PranabBardhan (1997).

Corruption has been defined in manydifferent ways, but “the most popularand simplest definition of corruption isthat it is the abuse of public power forprivate benefit” (Tanzi 1998, p. 8). Inthis definition, the private sector seemsto be excluded, which is of course notthe case; a more general definition is“that corruption is the intentional non-compliance with arm’s length relation-ship from this behavior for oneself orfor related individuals” (ib). There arevarious kinds of corruption, such asbribes to reduce costs; the literaturegives extensive analyses of factors thatstimulate corruption. Corruption issometimes involved in: satisfying regu-

lations and obtaining licenses to engagein particular activities (e.g. opening ashop; operating a taxi); land zoning andsimilar official decisions; access to pub-licly provided goods and services; deci-sions regarding procurement or publicinvestment contracts; control over theprovision of tax incentives; and hiringand promotion within the public sector.

The effects of corruption on the offi-cial economy can be seen from two dif-ferent perspectives: Paul Romer (1994)suggested that corruption, as a tax onex-post profits, may in general stimulatethe entry of new goods or technologywhich require an initial fixed-cost in-vestment. Paolo Mauro (1995) finds asignificant negative correlation betweena corruption index and the investmentrate or rate of GDP growth. A one-stan-dard-deviation improvement in the cor-ruption index is estimated by Mauro toincrease the investment rate by about3 percent. Johnson, Kaufmann, andZoido-Lobatón (1998b, p. 39) find a sig-nificant relationship between corrup-tion and GDP growth (an increase incorruption on an indexed scale from 0to 6 by one point decreases GDPgrowth by 0.84 percentage points) butthe relationship becomes insignificant ifthe shadow economy is entered as an in-dependent variable. In contrast, Bard-han (1997, p. 1329) concludes that “it isprobably correct to say that the processof economic growth ultimately gener-ates enough forces to reduce corrup-tion”—a view supported by Rose-Acker-mann (1997), who further argues thatany reform that increases the competi-tiveness of the economy will help re-duce incentives for corruption. Thus,policies that liberalize foreign trade andremove entry barriers for industry pro-mote competition and reduce corrup-tion. Such reforms will also encouragefirms to move from the shadow econ-omy into the official economy, where


they can obtain access to capital at mar-ket rates. Rose-Ackermann (1997, p. 21)concludes that “going underground is asubstitute for bribery, although some-times firms bribe officials in order toavoid the official states.”

Only a few studies empirically inves-tigate the relationship between theshadow economy and corruption, eitherin a country or over a sample of coun-tries. In their empirical investigation offorty-nine countries in Latin America,the OECD, and the post-communistcountries of Eastern Europe and theformer Soviet Union, Johnson, Kauf-mann, and Zoido-Lobatón (1998a, p. 21)find a statistically significant relation-ship between the various measures ofbribery or corruption and the shadoweconomy: ceteris paribus, a one-pointimprovement in the corruption indexICRG19 leads to an eight to eleven per-centage point decline in the shadoweconomy. Using another measure forcorruption, the transparency Interna-tional Corruption Index,20 Johnson,Kaufmann, and Zoido-Lobatón (1998b)found that, ceteris paribus, a one-pointincrease in this index decreases theshadow economy by 5.1 percentagepoints. Friedman, Johnson, Kaufmann,and Zoido-Lobatón (1999, p. 27) con-clude: “In summary, the relationshipbetween the share of the unofficialeconomy and rule of law (includingcorruption) is strong and consistentacross eight measures provided by sixdistinct organizations. All eight of theindices suggest that countries withmore corruption have a higher share ofthe unofficial economy.”

To summarize, the relationship be-

tween the size of the shadow economyand the amount of corruption is strongand consistent, as different measuresshow. Whereas Rose-Ackermann con-cludes from her work that going under-ground is a substitute for corruption(bribery), the empirical results ofJohnson, Kaufmann, and Zoido-Lobatón(1998b) point more to a complementaryprocess: Countries with more corrup-tion and bribery have larger shadoweconomies.

6. Methods of Estimating the Size of the Shadow Economy

The three methods most widely usedto measure the size and development ofthe shadow economy are discussed inthe following three subsections. Moredetailed discussions are given in Freyand Pommerehne (1984); Feige (1989);Thomas (1992, 1999); and Schneider(1986, 1994a, 1998a, 1999).

6.1 Direct Approaches

These are micro approaches that em-ploy either surveys and samples basedon voluntary replies, or tax auditing andother compliance methods. Sample sur-veys are widely used in a number ofcountries to measure the shadow econ-omy. The direct method of voluntarysample surveys was used for Norway byArne J. Isachsen, Jan Klovland, andSteinar Strom (1982); and Isachsen andStrom (1985). For Denmark thismethod was used by Mogensen, Kvist,Körmendi, and Pedersen (1995), whoestimate the shadow economy to be 2.7percent of GDP for 1989; 4.2 percentfor 1991; 3.0 percent for 1993; and 3.1percent for 1994. Further results forother countries can be found in Table 8.

The main advantage of this methodlies in the detailed information that canbe gained about the structure of theshadow economy. But results from

19 This index ranks between 1 and 6 (with 6meaning no corruption) and was averaged byJohnson, Kaufmann, and Zoido-Lobatón (1998b,p. 21) for the 1990s.

20 This index ranks between 0 and 10 (10 meansno corruption).


these surveys are sensitive to the waythe questionnaire is formulated, and, aswith all surveys, precision and resultsdepend greatly on the respondents’ will-ingness to cooperate. Most interviewedhesitate to confess fraudulent behavior,and quite often responses are unreli-able, making it difficult to estimate theextent of undeclared work. The advan-tages and disadvantages of this methodare extensively discussed by Mogensenet al. (1995).

Estimates of the shadow economy canalso be based on the discrepancy be-tween income declared for tax purposesand that measured by selective checks.Fiscal auditing programs have been par-ticularly effective in this regard. De-signed to measure the amount of unde-clared taxable income, they have beenused to calculate the shadow economyin several countries. For the United Statessee for example IRS (1979, 1983); C. P.Simon and A. D. Witte (1982); Witte(1987); Charles T. Clotfelter (1983);and Feige (1986). A detailed discussionis given in Bruno Dallago (1990) and inThomas (1992).

A number of difficulties beset this ap-proach. First, using tax compliance datais equivalent to using a (possibly biased)sample of the population. However,since in general selection of taxpayersfor audit is based on properties of sub-mitted (tax) returns which indicatesome likelihood of (tax) fraud, such asample is not a random one of thewhole population. This factor is likely tobias compliance-based estimates of theshadow economy. Second, estimatesbased on tax audits reflect only thatportion of shadow economy incomewhich the authorities succeeded in dis-covering, and this is likely to be only afraction of hidden income.

A further disadvantage of the two di-rect methods (surveys and tax auditing)is that they lead only to point estimates.

Moreover, it is unlikely that they cap-ture all shadow activities, so they can beseen as providing lower-bound esti-mates. They are unable (at least at pres-ent) to provide estimates of the devel-opment and growth of the shadoweconomy over a longer period of time.As already argued, they have at leastone considerable advantage. They canprovide detailed information aboutshadow economy activities and thestructure and composition of labor inthe shadow economy.

6.2 Indirect Approaches

These approaches, which are alsocalled indicator approaches, are mostlymacroeconomic, and use various eco-nomic and other indicators that containinformation about the developmentover time of the shadow economy. Cur-rently there are five such indicators,discussed next.

6.2.1 Discrepancy between National Expenditure and Income Statistics

In national accounting, the incomemeasure of GNP should be equal to theexpenditure measure of GNP. Thus, ifan independent estimate of the expen-diture side of the national accounts isavailable, the gap between the expendi-ture measure and the income measurecan be used as an indicator of theextent of the shadow economy.

This approach was used by A. Franz(1983) for Austria; by Kerrick MacAfee(1980), Michael O’Higgins (1989), andJames D. Smith (1985) for Great Brit-ain; by Hans-Georg Petersen (1982) andDaniela Del Boca (1981) for Germany;and by T. Park (1979) for the UnitedStates. The latest international compari-son of the shadow economy using mi-cro-level data was undertaken by TihoYoo and Jin K. Hyun (1998), who calcu-late the size of the shadow economiesof Korea (1996: 20.3 percent), Taiwan


(1995: 16.5 percent), Italy (1995: 19.2percent), Spain (1990: 50.5 percent),Russia (1995: 74.9 percent) and Hun-gary (1994: 56.9 percent). See Thomas(1992) for a survey and critical remarks.

Since national account statisticianswill be anxious to minimize this discrep-ancy, the initial discrepancy or first esti-mate, rather than the published dis-crepancy, should be employed for thispurpose. If all the components of theexpenditure side were measured with-out error, then this approach would in-deed yield a good estimate of the scaleof the shadow economy. However, thisis not the case, unfortunately, and thediscrepancy therefore reflects all omis-sions and errors everywhere in the na-tional account statistics as well as theshadow economy activity. These esti-mates are therefore of questionablereliability.21

6.2.2 Discrepancy between Official and Actual Labor Force

A decline in labor force participationin the official economy can be seen asan indication of increased activity in theshadow economy, if total labor forceparticipation is assumed to be constant,ceteris paribus. Such studies have beendone for Italy (Bruno Contini 1981,1982; Del Boca 1981) and for theUnited States (David M. O’Neill 1983).

The weakness of this method is thatdifferences in the rate of participationmay have other causes. Moreover, peo-ple can work in both the shadow and theofficial economies. Therefore, such esti-mates may be viewed as weak indicatorsof the size of the shadow economy.

6.2.3 The Transactions Approach

This approach was developed byFeige (1979, 1989, and 1996). Furtherapplication can be found for the Neth-erlands (Werner C. Boeschoten andMarcel M. G. Fase 1984), and for Ger-many (Enno Langfeldt 1984). Feige as-sumes that there is a constant relationover time between the volume of trans-actions and official GNP. This approachtherefore starts from Fisher’s quantityequation, MV = pT (with M = money,V = velocity, p = prices, and T = totaltransactions). Assumptions have to bemade about the velocity of money andthe relationships between the value oftotal transactions (pT) and total (= offi-cial + unofficial) nominal GNP. Relat-ing total nominal GNP to total transac-tions, the GNP of the shadow economyis calculated by subtracting officialGNP from total nominal GNP. How-ever, Feige has to assume a base yearin which there is no shadow economy,and therefore the ratio of pT to totalnominal (official = total) GNP wasnormal.

This method, too, has several weak-nesses; for instance, the assumption of abase year with no shadow economy, andthe assumption that the ratio of transac-tions to official GNP is constant overtime. Moreover, to obtain reliable es-timates, precise figures of the totalvolume of transactions need to be avail-able. This availability might be espe-cially difficult to achieve for cash trans-actions, because they depend, amongother factors, on the durability of banknotes, in terms of the quality of the pa-per on which they are printed. In thisapproach the additional assumption ismade that all variations in the ratio be-tween the total value of transactions andthe officially measured GNP are due tothe shadow economy. This means that aconsiderable amount of data is required

21 A related approach is pursued by C. Pis-sarides and G. Weber (1988), who use micro datafrom household budget surveys to estimate the ex-tent of income understatement by the self-em-ployed. In this micro approach, more or less thesame difficulties arise, and the figures calculatedfor the shadow economies seem to be crude.


in order to eliminate financial transac-tions from pure cross payments whichare legal and have nothing to do withthe shadow economy. For a detailedcriticism of the transaction approach,see Boeschoten and Fase (1984), Freyand Pommerehne (1984), Kirchgaessner(1984), Tanzi (1982, 1986), Dallago(1990), Thomas (1986, 1992, 1999) andGiles (1999a). In general, although thisapproach is theoretically attractive, theempirical requirements necessary to ob-tain reliable estimates are so difficult tofulfil that its application may lead todoubtful results.

6.2.4 The Currency Demand Approach

Phillip Cagan (1958) initiated thecurrency demand approach, correlatingcurrency demand and tax pressure forthe United States over the period 1919–55. Pierre M. Gutmann (1977) used asimilar approach but without statisticalprocedures, looking only at the ratio be-tween currency and demand depositsover the years 1937–76.

Cagan′s approach was further devel-oped by Tanzi (1980, 1983), who econo-metrically estimated a currency demandfunction for the United States for 1929–80. His approach assumes that shadowtransactions are undertaken in the formof cash payments, so as to leave notraces for the authorities. An increase inthe shadow economy will therefore in-crease the demand for currency. To iso-late the resulting excessive demand forcurrency, an equation for currency de-mand is econometrically estimated overtime. All conventional possible factors,such as development of income, pay-ment habits, interest rates, and so on,are controlled for. Additionally, suchvariables as the direct and indirect taxburden, government regulation, and thecomplexity of the tax system, which areassumed to be the major factors causingpeople to work in the shadow economy,

are included in the estimation equation.The basic regression equation for cur-rency demand proposed by Tanzi (1983)is:

ln(C/M2)t = βO + β1ln(1 + TW)t + β2

ln(WS/Y)t + β3lnRt + β4ln(Y/N) + ut

with β1 > 0, β2 > 0, β3 < 0, β4 > 0

where ln denotes natural logarithms; C/M2is the ratio of cash holdings to currentand deposit accounts; TW is a weightedaverage tax rate (to proxy changes in thesize of the shadow economy); WS/Y is aproportion of wages and salaries in na-tional income (to capture changing pay-ment and money holding patterns); R isthe interest paid on savings deposits (tocapture the opportunity cost of holdingcash); and Y/N is the per capita income.

The excessive increase in currency—the amount unexplained by the conven-tional factors mentioned above—is thenattributed to the rising tax burden andother factors leading people to work inthe shadow economy. Figures for thesize and development of the shadoweconomy can be initially calculated bycomparing the development of currencywhen taxes and government regulationsare at their lowest values, with the de-velopment of currency at the current(higher) levels of taxation and regula-tions. Next, assuming the same incomevelocity for currency in the shadoweconomy as for money (as measured byM1) in the official economy, the size ofthe shadow economy can be computedand compared to the official GDP. Thiscurrency demand equation is criticizedby Thomas (1999), and some of the criti-cisms are addressed by Giles (1999a,b)and Bhattacharyya (1999), who use thelatest econometric techniques.

The currency demand approach isone of the most commonly used meth-ods. It has been applied to many OECD


countries (e.g. Schneider 1997, 1998a;Johnson, Kaufmann, and Zoido-Lobatón(1998a); and Colin C. Williams and JanWindebank (1995). But it has neverthe-less been criticized on various grounds(e.g. Thomas 1986, 1992, 1999; Feige1986; and Pozo 1996). The most com-monly raised objections to this methodare:

(i) Not all transactions in the shadoweconomy are paid in cash. Isachsenand Strom (1980, 1985), using thesurvey method, concluded that inNorway in 1980, roughly 80 percentof all transactions in the under-ground sector were paid in cash.The size of the total shadow econ-omy (including barter) may thus beeven larger than previously estimated.

(ii) Most studies consider only one par-ticular factor, the tax burden, as acause of the shadow economy. Butothers (such as the impact of regu-lation, taxpayers’ attitudes towardthe state, “tax morality,” and so on)are not considered, because reliabledata for most countries are notavailable. If, as seems likely, theseother factors also affect the extentof the hidden economy, it mightagain be higher than reported inmost studies.22

(iii) Increases in currency demand de-posits are due largely to a slowdownin demand deposits, rather than to

an increase in currency caused byactivities in the shadow economy, atleast in the United States, as dis-cussed by Gillian Garcia (1978),Park (1979), and Feige (1996). Also,Derek Blades (1982) and Feige(1986, 1997) criticize Tanzi’s stud-ies on the grounds that the US dol-lar is used as an international cur-rency. Tanzi should have considered(and controlled for) the US dollars,which are used as an internationalcurrency and held in cash abroad.23

Frey and Pommerehne (1984) andThomas (1986, 1992, 1999) claimthat Tanzi’s parameter estimates arenot very stable.24

(iv) Another weak point of this proce-dure, in most studies, is the as-sumption of the same velocity ofmoney in both types of economies.As Hill and Kabir (1996) for Canadaand Klovland (1984) for the Scandi-navian countries argue, there isalready considerable uncertaintyabout the velocity of money in theofficial economy; the velocity ofmoney in the hidden sector is evenmore difficult to estimate. Withoutknowledge about the velocity of

22 One (weak) justification for the use of onlythe tax variable is that this variable has a verystrong impact on the size of the shadow economyin the studies known to the authors. One excep-tion is the study by Frey and Weck-Hannemann(1984) where the variable “tax immorality” has aquantitatively larger and statistically stronger in-fluence than the direct tax share in the model ap-proach. In the study by Pommerehne andSchneider (1985), for the U.S., besides various taxmeasures, data for regulation, tax immorality,minimum wage rates are available, the tax variablehas a dominating influence and contributesroughly 60–70 percent of the size of the shadoweconomy. See also Zilberfarb (1986).

23 In another study by Tanzi (1982, especiallypp. 110–13), he explicitly deals with this criticism.A very careful investigation of the amount of USdollars used abroad and in the shadow economyand “classical” crime activities was undertaken byKenneth Rogoff (1998), who concludes that largedenomination bills are a major driving force forthe growth of the shadow economy and classicalcrime activities due to reduced transactions costs.

24 However, in studies for European countries,Kirchgaessner (1983, 1984) and Schneider (1986)conclude that the estimation results for Germany,Denmark, Norway and Sweden are quite robustwhen using the currency demand method. Hilland Kabir (1996) find for Canada that the rise ofthe shadow economy varies with respect to the taxvariable used; they conclude “when the theo-retically best tax rates are selected and a range ofplausible velocity values is used, this method esti-mates underground economic growth between1964 and 1995 at between 3 and 11 percent ofGDP.” (Hill and Kabir 1996, p. 1553).


currency in the shadow economy,one has to accept the assumption ofan “equal” money velocity in bothsectors.

(v) Finally, the assumption of noshadow economy in a base year isproblematic, and relaxing the as-sumption would again imply an up-ward adjustment of the figures at-tained in the bulk of the studiesalready undertaken.

6.2.5 The Physical Input (Electricity Consumption) Method

The Kaufmann–Kaliberda Method.This method was used earlier by C.Lizzeri (1979), and Del Boca andFrancesco Forte (1982); and then la-ter by Alejandro Portes (1996); Kauf-mann and Aleksander Kaliberda (1996);and Johnson, Kaufmann, and Shleifer(1997). For a critique see Mária Lackó(1996, 1997, 1998). To measure overall(official and unofficial) economic activ-ity in an economy, Kaufmann and Ka-liberda (1996) assume that electricityconsumption is the single best physicalindicator of overall economic activity.Overall (official and unofficial) eco-nomic activity and electricity consump-tion have been empirically observedthroughout the world to move in lock-step, with an electricity/GDP elasticityusually close to one. By having a proxymeasurement for the overall economyand subtracting it from estimates of of-ficial GDP, Kaufmann and Kaliberdaderive an estimate of unofficial GDP.The difference between the growth of of-ficial GDP and the growth of electricityconsumption is attributed to the growthof the shadow economy. This method issimple and appealing; however, it canalso be criticized:

(i) Not all shadow economy activitiesrequire a considerable amount ofelectricity (e.g. personal services),

and other energy sources can beused (gas, oil, coal, etc.), so thatonly a part of the shadow economywill be captured.

(ii) Over time, there has been consider-able technical progress. The use ofelectricity is more efficient than inthe past, in both official and unoffi-cial uses.

(iii) There may be considerable differ-ences in the elasticity of electric-ity/GDP across countries or changesover time.25

The Lackó Method. Lackó (1996, 1998,1999) assumes that a certain part of theshadow economy is associated with thehousehold consumption of electricity, in-cluding so-called household production,do-it-yourself activities, and other non-registered production and services. Lackóassumes that in countries where the partof the shadow economy associated withhousehold electricity consumption ishigh, the rest of the hidden economy—that is, the part Lackó cannot measure—will also be high. Lackó (1996, pp. 19 ff.)assumes that in each country a part ofthe household consumption of electricityis used in the shadow economy.

Lackó’s approach (1998, p. 133) can bedescribed by the following two equations:

lnEi = α1lnCi + α2lnPRi + α3Gi

+ α4Qi + α5Hi + ui (1)with

α1 > 0, α2 < 0, α3 > 0, α4 < 0, α5 > 0Hi = β1 Ti + β2 (Si − Ti) + β3 Di (2)

with β1 > 0, β2 < 0, β3 > 0

where i is the number assigned to thecountry; Ei is per capita household elec-tricity consumption in country i in Mtoe;Ci is per capita real consumption ofhouseholds without the consumption of

25 Johnson, Kaufmann, and Shleifer (1997)attempt to adjust for changes in the elasticity.


electricity in country i in US dollars (atpurchasing power parity); PRi is the realprice of consumption of 1 kwh of resi-dential electricity in US dollars (at pur-chasing power parity); Gi is the relativefrequency of months with the need toheat houses in country i; Qi is the ratioof energy sources other than electricityto all energy sources in household en-ergy consumption; Hi is the per capitaoutput of the hidden economy; Ti is theratio of the sum of paid personal income,corporate profit, and taxes on goods andservices to GDP; Si is the ratio of publicsocial welfare expenditures to GDP; andDi is the sum of the numbers of depen-dants over 14 years and of inactive earners,both per 100 active earners.

In a cross-country study, Lackó esti-mates equation (1) substituting Hi byequation (2). The econometric resultscan then be used to order the countrieswith respect to electricity use in theirshadow economies. For the calculationof the actual size (value added) of theshadow economy, Lackó needs to knowhow much GDP is produced by one unitof electricity in the shadow economy ofeach country. Since these data are notknown, she takes one of the shadoweconomy estimates obtained using an-other approach, and applies this propor-tion to the other countries. Lackó usesthe shadow economy of the UnitedStates in the early 1990s as such a base(the shadow economy value of 10.5 per-cent of GDP taken from B. Morris1993), and then calculates the size ofthe shadow economy for other coun-tries. Lackó’s method is also open tocriticism:

(i) Not all shadow economy activi-ties require a considerable amountof electricity, and other energysources can be used.

(ii) Shadow economy activities do not takeplace only in the household sector.

(iii) It is doubtful whether the ratio ofsocial welfare expenditures can beused as the explanatory factor forthe shadow economy, especially intransition and developing countries.

(iv) It is unclear which is the mostreliable base value of the shadoweconomy to calculate its size for allother countries, especially for thetransition and developing countries.

6.3 The Model Approach

The pioneers of this approach areWeck (1983), Frey and Weck (1983a,b),Frey and Weck-Hannemann (1984),who applied it to cross-section datafrom the twenty-four OECD countriesfor various years. Before turning tothis approach they developed the con-cept of “soft modeling” (Frey, Weck,and Pommerehne 1982; Frey and Weck1983a,b), an approach that has beenused to provide a ranking of the relativesize of the shadow economy in differentcountries.

All methods described so far for esti-mating the size and development ofthe shadow economy consider just oneindicator of all effects of the shadoweconomy. However, its effects showup simultaneously in the production, la-bor, and money markets. A more impor-tant critique is that some of the mone-tary approach studies consider just onecause, the burden of taxation. Themodel approach explicitly considers themultiple causes of, as well as the multi-ple effects of, the shadow economy.The empirical method, quite differentfrom those discussed so far, is basedon the statistical theory of unobservedvariables, which considers multiplecauses and multiple indicators of thephenomenon. A factor-analytic ap-proach is used to measure the hiddeneconomy as an unobserved variable overtime. The unknown coefficients are es-timated in a set of structural equations


within which the unobserved variablecannot be measured directly (see Den-nis Aigner, Schneider, and VictorGhosh 1988).

The dynamic multiple-indicators mul-tiple-causes model consists of two parts.The measurement model links the un-observed variables to observed indica-tors. The structural equations modelspecifies causal relationships among theunobserved variables. In this case, thereis one unobserved variable, the size ofthe shadow economy. It is assumed tobe influenced by a set of indicators ofthe shadow economy’s size, thus captur-ing the structural dependence of theshadow economy on variables that maybe useful in predicting its movementand size in the future. The interactionover time between the causes Zit (i = 1,2, . . . , k) the size of the shadow econ-omy Xt, and the indicators Yjt (j = 1,2, . . . , p) is shown in Figure 1.

As discussed in Section 3 above, thereare three main possible causes of theshadow economy: high taxation, heavyregulation, and declining “tax morality”(citizens’ attitudes toward the state),which describes the readiness of indi-viduals (at least partly) to leave their of-ficial occupations and enter the shadow

economy. When applying this approachfor European countries, Frey andWeck-Hannemann (1984) had difficultyobtaining reliable data for regulationand tax morality. Their study was criti-cized by Claus Helberger and HansKnepel (1988), who argue that the re-sults were unstable with respect tochanging variables in the model andover time.

Indicators. A change in the size ofthe shadow economy may be reflectedin the following indicators: monetary in-dicators—if activities in the shadoweconomy rise, additional monetarytransactions are required; labor mar-ket—increasing participation of work-ers in the hidden sector results indecreased participation in the officialeconomy (similarly, increased activitiesin the hidden sector may be reflected inshorter working hours in the officialeconomy); production market—an in-crease in the shadow economy meansthat inputs (especially labor) move outof the official economy (at least partly),and this displacement might have a de-pressing effect on the official growthrate of the economy.

The model approach has been furtherdeveloped by Giles (1999a,b) and Giles,


Linsey M. Tedds, and Gugsa Werkneh(1999), who estimate a comprehensivedynamic multiple-indicators multiple-causes model to get a time-series indexof the hidden/measured output of NewZealand and Canada, and then estimatea separate “cash-demand model” to ob-tain a benchmark for converting thisindex into percentage units. Unlike ear-lier empirical studies, they paid atten-tion to the nonstationarity and possiblecointegration of time-series data in bothmodels. This model treats hidden out-put as a latent variable, and uses several(measurable) causal variables and indi-cator variables. The former include theaverage and marginal tax rates, infla-tion, real income, and the degree ofregulation in the economy. The latterinclude changes in the (male) laborforce participation rate and in thecash/money supply ratio. In their cash-demand equation they allow for differ-ent velocities of currency circulation inthe hidden and recorded economies.Their cash-demand equation is not usedas an input to determine the variationin the hidden economy over time; it isused only to obtain the long-run aver-age value of hidden/measured output,so that the ratio predicted by the modelcan be used to estimate the shadoweconomy.

7. The Empirical Findings in More Detail

7.1 How Large Is the Shadow Economy?

For single countries, and sometimesfor a group of countries (like the OECDor transition countries), the size of theshadow economy has been estimated us-ing various methods and different timeperiods, but until now there has beenno consistent comparison of estimatesof the size of the shadow economies ofvarious countries, for a fixed period,

generated by using similar methods. InTables 4 to 6, such a comparison ismade, reporting the results for theshadow economies of 76 countries for1989–90 and 1990–93 using the physicalinput (electricity) method, the currencydemand approach and the model ap-proach. Unfortunately, comparison ofthe size of shadow economies betweencountries remains crude, since at leasttwo methods have not been applied forall seventy-six countries.26

7.1.1 Developing Countries

Table 4 shows the results of applyingthe physical input (electricity), currencydemand, and model approaches for de-veloping countries. The results fromeight countries in Africa are reported.Among these, Nigeria and Egypt havethe largest shadow economies with 76percent and 68 percent of GDP; Mauri-tius has the smallest shadow economywith 20 percent. Applying the currencydemand approach, Tanzania had ashadow economy of 31 percent (ofGDP) in 1989–90, and South Africa, 9percent in 1989–90. The ranking of thesize of the shadow economies for theAfrican countries is supported by simi-lar findings and anecdotal evidencefrom Lubell (1991); Lawrence Chicker-ing and Muhamed Salahdine (1991);and Pozo (1996).

For Central and South Americancountries, we have two estimates—oneusing the physical input method (Lackó1996) and one the model approach(Loayza 1996). For some countries, theestimates of the size of the shadoweconomy are quite similar, e.g., Vene-zuela, Brazil, and Guatemala. For oth-ers there are great differences, e.g.,

26 In this comparison the same time periods(1989–90 or 1990–93) are used for all countries,and, if possible, the values were calculated asaverages over the time periods.


TABLE 4SHADOW ECONOMIES IN DEVELOPING COUNTRIES

Size of Shadow Economy (as % of GDP)

Physical InputMethod

Currency DemandApproach MIMIC Approach

Countries Average 1989–90 Average 1989–90 Average 1990–93

Africa Botswana 27.0 — — Egypt 68.0 — — Mauritius 20.0 — — Morocco 39.0 — — Nigeria 76.0 — — South Africa — 9.01 — Tanzania — 31.02 — Tunisia 45.0 — —

Central and South America Argentina — — 21.8 Bolivia — — 65.6 Brazil 29.0 — 37.8 Chile 37.0 — 18.2 Colombia 25.0 — 35.1 Costa Rica 34.0 — 23.2 Ecuador — — 31.2 Guatemala 61.0 — 50.4 Honduras — — 46.7 Mexico 49.0 33.03 27.1 (35.1)3

Panama 40.0 — 62.1 Paraguay 27.0 — — Peru 44.0 — 57.4 Uruguay 35.2 — — Venezuela 30.0 — 30.8

Asia Cyprus 21.0 — — Hong Kong 13.0 — — India — 22.44 — Israel 29.0 — — Malaysia 39.0 — — Philippines 50.0 — — Singapore 13.0 — — South Korea 38.0 — 20.35

Sri Lanka 40.0 — — Taiwan — — 16.55

Thailand 71.0 — —

Sources: Authors’ calculations using values for developing countries in Africa and Asia from Lackó (1996, Table 18);for Central and South America from Loayza (1996). A dash means no value available. Other sources:1 For South Africa, G. M. Hartzenburg and A. Leimann (1992); they used the currency demand approach.2 For Tanzania, M. S. D. Bagachwa and A. Naho (1995, p. 1394); they used the currency demand approach.3 For Mexico, Pozo (1996) estimates 33.0% (1989–90) and 35.1% (1990–93) using the currency demand approach.4 Authors’ calculations using the absolute figures of Bhattacharyya (1999).5 For Taiwan, Yoo and Hyun (1998) used the income discrepancy method; also for South Korea for 1990–93.


Panama, Peru, and Mexico. Using themodel approach for a ranking of theSouth American countries, the biggestshadow economies for the period 1990–93 were in Bolivia with 65.6 percent ofGDP; Panama with 62.1 percent; Peruwith 57.4 percent; and Guatemala with50.4 percent. The smallest can be foundin Costa Rica with 23.2 percent; Argen-tina with 21.8 percent; and Chile with18.2 percent. This ranking for Central

and South America is also supportedby similar findings in Pozo (1996), Lip-pert and Walker (1997), and Lubbel(1991). For Mexico, the results from allthree methods are shown. Whereas themodel approach and the currency de-mand method are in a similar range(27.1 percent and 35.1 percent), thephysical input method provides a sizeof 49 percent, far above the othertwo.

TABLE 5SHADOW ECONOMIES IN TRANSITION COUNTRIES

Size of Shadow Economy (as % of GDP) Physical Input Method using values from Johnson et al.∗ and from Lackó

Average1989–90

Average1990–93

Average1994–95

Countries Johnson Lackó Johnson Lackó Johnson Lackó

Former Soviet Union1

Azerbaijan 21.9 — 33.8 41.0 59.3 49.1 Belarus 15.4 — 14.0 31.7 19.1 45.4 Estonia 19.9 19.5 23.9 35.9 18.5 37.0 Georgia 24.9 — 43.6 50.8 63.0 62.1 Kazakhstan 17.0 13.0 22.2 29.8 34.2 38.2 Kyrgyzstan — 13.9 — 27.1 — 35.7 Latvia 12.8 18.4 24.3 32.2 34.8 43.4 Lithuania 11.3 19.0 26.0 38.1 25.2 47.0 Moldavia 18.1 — 29.1 — 37.7 — Russia 14.7 — 27.0 36.9 41.0 39.2 Ukraine 16.3 — 28.4 37.5 47.3 53.7 Uzbekistan 11.4 13.9 10.3 23.3 8.0 29.5Average 16.7 16.2 25.7 34.9 35.3 43.6

Central and Eastern Europe Bulgaria 24.0 26.1 26.3 32.7 32.7 35.0 Croatia 22.82 — 23.52 39.0 28.52 38.2 Czech Republic 6.4 23.0 13.4 28.7 14.5 23.2 Hungary 27.5 25.1 30.7 30.9 28.4 30.5 Macedonia — — — 40.4 — 46.5 Poland 17.7 27.2 20.3 31.8 13.9 25.9 Romania 18.0 20.9 16.0 29.0 18.3 31.3 Slovakia 6.9 23.0 14.2 30.6 10.2 30.2 Slovenia — 26.8 — 28.5 — 24.0Average 17.6 17.6 20.6 32.4 20.9 31.6

Sources: ∗ Authors’ calculations using values from Johnson, Kaufmann, and Shleifer (1997, Table 1, p. 182–83), andJohnson, Kaufmann, and Zoido-Lobatón (1998a, p. 351). Lackó values from Lackó (1999, Table 8).1 For the former Soviet Union states in the column 1989/90 only data for 1990 was available from Johnson,Kaufmann and Shleifer (1997).2 For Croatia see Sanja Madzarevic and Davor Milkulic (1997, Table 9, p. 17); they used the discrepancy method.


In Asia, Thailand ranks first with 71.0percent, followed by the Philippineswith 50 percent and Sri Lanka with 40percent. Hong Kong and Singaporerank lowest with a shadow economy of13 percent of GNP.

7.1.2 Transition Countries

The physical input method was usedfor the transition countries in Centraland Eastern Europe and the states ofthe former Soviet Union. The resultsare shown in Table 5; they cover the pe-

riods 1989–90, 1990–93 and 1994–95.27

Considering the physical input methodby Johnson, Kaufman, and Shleifer(1997) (and respectively, the Lackó val-ues) and the countries of the former So-viet Union over the period 1990–93,28

Georgia has the largest shadow economy

TABLE 6SHADOW ECONOMIES IN OECD COUNTRIES

Size of Shadow Economy (as % of GDP) using:

Countries

Physical Input(Electricity)

Method 1990

CurrencyDemandMethod,

Schneiderfigures

Average 1989–90

CurrencyDemandMethod,

Schneiderfigures

Average 1990–93

CurrencyDemand Method,

Johnson et al.figures

Average 1990–93

Australia 15.3 10.1 13.0 13.1Austria 15.5 5.1 6.1 5.8Belgium 19.8 19.3 20.8 15.3Canada 11.7 12.8 13.5 10.0Denmark 16.9 10.8 15.0 9.4Finland 13.3 — — —France 12.3 9.0 13.8 10.4Germany 14.6 11.8 12.5 10.5Great Britain 13.1 9.6 11.2 7.2Greece 21.8 — — 27.2Ireland 20.6 11.0 14.2 7.8Italy 19.6 22.8 24.0 20.4Japan 13.2 — — 8.5Netherlands 13.4 11.9 12.7 11.8New Zealand1 — 9.2 9.0 9.0Norway 9.3 14.8 16.7 5.9Portugal 16.8 — — 15.6Spain2 22.9 16.1 17.3 16.1Sweden 11.0 15.8 17.0 10.6Switzerland 10.2 6.7 6.9 6.9USA 10.5 6.7 8.2 13.9Average 15.1 11.9 13.5 11.3

Sources: Physical input method, Lackó (1996, 1997, 1998, 1999); currency demand approach, Schneider (1994a,1998a), Johnson, Kaufmann, and Zoido-Lobatón (1998a,b), and Williams and Windebank (1995).1 Calculated using the MIMIC method and currency demand approach. Source Giles (1999b).2 Calculated from Ignacio Mauleon (1998).

27 For the first period 1989–90 the results canonly be seen as very crude ones, because the col-lapse of the communist regimes took place in theyears 1989 and 1990.

28 The period 1989–90 is not discussed here be-cause in this period the former Soviet Union wasbreaking up.


with 43.6 (50.8) percent of GDP, fol-lowed by Azerbaijan with 33.8 (41.0)percent and Moldova 29.1 percent. Rus-sia is in the middle with a shadow econ-omy of 27 (36.9) percent. According tothe Johnson, Kaufmann, and Zoido-Lo-batón (1998b) figures, Belarus with 14percent and Uzbekistan with 10.3 per-cent have the smallest values. ExceptUzbekistan (only for the Johnson fig-ures) all other former Soviet Unioncountries experienced a strong increasein their shadow economies from anaverage of 25.7 percent (Lackó value:34.9 percent) for 1990–93, to 35.3 per-cent (Lackó value: 43.6 percent) for1994–95, calculated over all twelvecountries of the former Soviet Union. Amore detailed analysis of the situationin the Ukraine is given by Kaufmann(1997).

Turning to the transition countries ofCentral and Eastern Europe, and con-sidering the period 1990–93 and theJohnson, Kaufmann, and Zoido-Lobatón(1998b) figures, Hungary has the larg-est shadow economy with 30.7 percentof GNP, followed by Bulgaria with 26.3percent. The lowest two are the CzechRepublic with 13.4 percent and Slovakiawith 14.2 percent. Considering theLackó figures, Macedonia has the larg-est shadow economy with 40.4 percent,followed by Croatia with 39.0 percent.Whereas for the former Soviet Unioncountries a strong increase over the twoperiods 1990–93 and 1994–95 was ob-served, the average size of the shadoweconomy of Central and Eastern Euro-pean states was almost stable over thesetwo periods. The Johnson, Kaufmann,and Zoido-Lobatón (1998b) figuresshow an average shadow economy of theCentral and Eastern European states of20.6 percent (Lackó 32.4) over 1990–93; and over the period 1994–95Johnson, Kaufmann, and Zoido-Lobatón(1998b) show an average size for the

Central and Eastern European states of20.9 percent (Lackó 31.6).

Lackó estimates larger shadow econo-mies for the transition countries than doJohnson, Kaufmann, and Zoido-Lobatón(1998b), perhaps because Lackó uses anestimate of household electricity con-sumption, whereas Johnson, Kaufmann,and Zoido-Lobatón use overall electricityconsumption.

7.1.3 OECD Countries

For the twenty-one OECD western-type countries, either the currency de-mand method or the physical inputmethod were used. For the currencydemand method, two series of figuresare shown—one from Schneider and onefrom Johnson, Kaufmann, and Zoido-Lobatón (1998a,b). The main differencebetween the two is that Johnson, Kauf-mann, and Zoido-Lobatón (1998a,b) useaverage values of the size of the shadoweconomy of a country coming from dif-ferent sources, if a monetary approachwas applied, whereas in Schneider thecurrency-demand approach is used forthese countries and only one value forthat year (or an average over a time pe-riod) is used. The problem using aver-ages from various sources is (a) that thetime period is greater (1985–95); and(b) the specification of the monetaryapproaches from different authors maybe quite different.

Considering the period 1990–93 andusing the series by Johnson, Kaufmann,and Zoido-Lobatón (Table 6), where es-timates of the shadow economy for mostOECD countries are available (20 outof the 21 investigated countries), thesouthern European countries have thelargest shadow economies: Greece (27.2percent), Italy (20.4 percent), Spain(16.1 percent), and Portugal (15.6 per-cent). A similar result can be foundwhen using figures of Schneider, and toa much lesser extent the ones achieved


by the physical input (electricity)method by Lackó (1997). At the lowerend, Johnson, Kaufmann, and Zoido-Lo-batón rank Switzerland (6.9 percent),Norway (5.9 percent), and Austria (5.8percent); whereas Schneider finds theUSA (8.2 percent), Switzerland (6.9percent), and Austria (6.1 percent). Ingeneral, this ranking of the size of theshadow economies of the OECD coun-tries calculated by Schneider is supportedby other studies. Frey and Pommerehne(1984), Frey and Weck-Hannemann(1984), Williams and Windebank (1995),Thomas (1992), and Lippert and Walker(1997) reach quite similar rankings.

In Table 7, the latest results are shown

for OECD countries over the period1994–95, and for the period 1996–97. Theranking of the sizes of the shadow econo-mies of the results are similar to theones in Table 6. However, the shadoweconomy has increased compared to1990–93 in all OECD countries.Whereas the average size of the shadoweconomy of the investigated OECDcountries was 13.5 percent of the GDPin 1990–93, this value increased to 16.0percent of GDP in the years 1994–95. Afurther increase can be observed for theinvestigated OECD countries to 16.9percent for the period 1996–97. Even inthe late 1990s, the shadow economy isstill growing in most OECD countries.

7.2 Comparing the Results of the Different Methods

As discussed in Section 6, there are atleast nine different methods used to es-timate the shadow economy. In Table 8,the empirical results of the methods ap-plied to Canada, Germany, Great Britain,Italy and the United States are shown.

The survey method, which was usedfor all five countries, provides lower-bound estimates ranging from 1.5 per-cent to 4.5 percent for the period 1970–80. The tax auditing method provideshigher estimates, ranging from 2.9 per-cent to 8.2 percent for 1970–80. Bothmethods also show that the shadoweconomy increases over time (e.g. forthe United States). The two discrepancymethods (expenditure versus incomeand official versus actual labor force)show no clear pattern. For some coun-tries they produce high shadow econ-omy values (compared to the othermethods for these countries, e.g. Ger-many); for some low (e.g. Canada).They do not show a consistent time pat-tern. The physical input (electricity)method, for which only values for 1986–90 are available for all five countries,shows values in the middle size range

TABLE 7SHADOW ECONOMIES OF OECD COUNTRIES

1994–97

Size of Shadow Economy as % of GDPCurrency Demand Approach

Average 1994–95

Average 1996–97

Australia 13.8 13.9Austria 7.0 8.6Belgium 21.5 22.2Canada 14.8 14.9Denmark 17.8 18.2France 14.5 14.8Germany 13.5 14.8Great Britain 12.5 13.0Greece 29.6 30.1Ireland 15.4 16.0Italy 26.0 27.2Japan 10.6 11.3Netherlands 13.7 13.8New Zealand 11.31 —Norway 18.2 19.4Portugal 22.1 22.8Spain 22.4 23.0Sweden 18.6 19.5Switzerland 6.7 7.8USA 9.2 8.8Average 16.0 16.9

Sources: Authors’ calculations using the data bySchneider (1998a) and Schneider and Pöll (1999).1 1994 only, source Giles (1999b).


for all countries (average value of 12.7percent over all countries and all peri-ods). If one compares the three mone-tary approaches (currency demand,cash-deposit ratio, and transactions ap-proach), a clear pattern appears. Thelargest size of the shadow economiesfor all five countries resulted using thetransactions approach (Feige method)ranging from 15 to 35 percent of GNP(average value of 21.9 percent over allcountries and periods). Somewhat lowerresults are achieved with the cash-de-posit ratio (Gutmann method), rangingbetween 10 percent and 30 percent forall countries (average value of 15.5 per-cent over all countries and all periods).

Considerably lower values wereachieved using the currency demand ap-proach, ranging from 4 percent to 20percent of GNP over the period 1970–90 for all five countries (average valueof 8.9 percent over all countries and pe-riods). The currency demand approachshows a strongly rising shadow economyin all five countries, a result oppositethat given by the transactions and cashdeposit methods. The model approachshows values in the medium range from6.1 percent to 10.5 percent for the pe-riod 1976–80 (average value of 7.9 per-cent for all countries over all periods).In general, these results demonstratewhat a huge range of estimates of theshadow economy for a country in agiven time span are achievable usingdifferent calculation methods. Henceone should be very careful when inter-preting the size of the shadow economyin a country using only one method.

7.3 The Shadow Economy Labor Force

We now discuss the labor market inthe shadow economy.29 On the official

labor market, the costs that firms (andindividuals) have to pay when “offi-cially” hiring someone are tremendouslyincreased by the burden of tax and so-cial security contributions on wages, aswell as the legal administrative regula-tion to control economic activity. Invarious OECD countries, these costsare greater than the wage effectivelyearned by the worker—providing astrong incentive to work in the shadoweconomy. This is especially true inEurope (e.g. in Germany and Austria),where the total tax and social securityburden adds up to 100 percent on topof the wage effectively earned (see sec-tion 4.2, and for Italy see Dallago 1985,1990).

Working in the shadow economy mayconsist of a second job after (or evenduring) regular working hours; a secondform is work by individuals who do notparticipate in the official labor market;a third form is work by people (e.g.clandestine, social fraud, or illegal im-migrants) who are not allowed to workin the official economy.

The few existing results on theshadow economy labor force are shownin Table 9,30 which provides rough esti-mates of the size of the labor force inthe shadow economy for some OECDcountries. The estimations are basedeither on the survey or discrepancymethod (e.g. for Denmark, Italy,France) or on a calculation using thevalue added of the shadow economies,subtracting all material inputs and as-suming certain average values of earn-ings paid per hour. The results for Den-mark show that the population of adultDanes engaged in the shadow economyranged from 8.3 percent (of the total la-bor force) in 1980 to 22.5 percent in

29 Work in this area has been done by L. Frey(1972, 1975, 1978, 1980); M. A. Cappiello (1986);Lubell (1991); Pozo (1996); Bruce Bartlett (1998);and Tanzi (1999).

30 For developing countries, some literatureabout the shadow labor market exists, e.g. Dallago(1990), Pozo (1996), Loayza (1996), and especiallyChickering and Salahdine (1991).


1998. In Germany, this figure rose from8–12 percent in 1974–82 to 22 percentin 1998. Both countries, then, show astrong increase.

In other countries, the shadow econ-omy labor force is also quite large: inItaly 30–48 percent (1997–98); Spain11.5–32.3 percent (1997–98); Sweden19.8 percent (1997); and France 6–12percent (1997–98). In the EuropeanUnion at least 20 million workers andin OECD countries about 35 million(1997–98) work in the unofficial econ-omy. Moreover, the amount doubledwithin twenty years. The labor marketin the shadow economy is lively andmay provide one explanation for suchhigh and persistent unemploymentmeasured in many countries.

8. Summary and Conclusions

Many obstacles must be overcome tomeasure the size of the shadow econ-omy and to analyze its consequences onthe official economy, although someprogress has been made. In this surveywe have shown that although it is diffi-cult to estimate the size of the shadoweconomy, it is not impossible. We havedemonstrated that with the variousmethods—the currency demand, thephysical input measure, and the modelapproach—some insights can be pro-vided into the size and development ofthe shadow economy of the developing,transition, and OECD countries. Thereis no “best” or commonly acceptedmethod. Each approach has its specific

TABLE 8A COMPARISON OF ESTIMATES OF SHADOW ECONOMIES OF 5 OECD COUNTRIES USING 9 METHODS

Canada average over Germany average over Great Britain average over

Method1970–75

1976–80

1981–85

1986–90

1970–75

1976–80

1981–85

1986–90

1970–75

1976–80

1981–85

1986–90

Surveys of households — — 1.3 1.4 3.6 — — — 1.5 — — —Tax auditing — — 2.9 — — — — — — — — —Discrepancy bet. expenditure and income — — — — 11.0 10.2 13.4 — 2.5 3.6 4.2 —Discrepancy bet. officialand actual employment — — — — 23.0 38.5 34.0 — — — — —Physical input — — — 11.2 — — — 14.5 — — — 13.2Currency demand(Tanzi) 5.1 6.3 8.8 12.0 4.5 7.8 9.2 11.3 4.3 7.9 8.5 9.7Cash deposit ratio (Gutmann) 13.8 15.9 11.2 18.4 — — — — 14.0 7.2 6.2 —Transactions (Feige) — 26.5 15.4 21.2 17.2 22.3 29.3 31.4 17.2 12.6 15.9 —MIMIC (Frey andWeck-Hannemann) — 8.7 — — 5.8 6.1 8.2 — — 8.0 — —Number of methodsused 2 4 5 5 6 5 5 3 5 5 4 2

Notes: Values were grouped (when possible, averaged) in the time periods in order to undertake a roughcomparison.Sources: Authors’ calculations using the following sources:1 For Canada: Lippert and Walker (1997), Thomas (1992), Hill and Kabir (1996), Schneider (1997), and Jacques Bendelac and Pierre-Maurice Clair (1993).2 For Germany: Lippert and Walker (1997), Schneider (1994a,b) and Schneider (1997).3 Great Britain: Thomas (1992), Lippert and Walker (1997), Schneider (1994a,b, 1997), Pozo (1996).


strengths and weaknesses and can pro-vide specific insights and results. Thegeneral impression from the results ofthese estimates is that, for all countriesinvestigated, the shadow economy hasreached a remarkably large size. Al-though the different methods provide arather wide range of estimates, there isa common finding that the shadoweconomies of most transition and all in-vestigated OECD countries have beengrowing over the past decade. The samecan be said for the labor market in theshadow economy, which is attracting grow-ing attention due to high unemploymentin European OECD countries.

The analysis of causes shows that anincreasing burden of taxation and socialsecurity payments, combined with rising

state regulatory activities and labormarket restrictions (e.g. forced reduc-tion in working hours), are the majordriving forces for the size and growth ofthe shadow economy. But an interdisci-plinary approach seems to be necessaryfor a more comprehensive analysis,which would consider aspects like taxmorale, perceived fairness of the taxsystem, and institutional aspects as well.

The results on the shadow economy’seffects on the official economy (e.g. theofficial growth rate and tax revenue) areambiguous. According to some studies,a growing shadow economy has a nega-tive impact on official GDP growth. Butother studies show the opposite effect.Hence, it is important to undertake fur-ther research to gain more precise

TABLE 8 (Cont.)

Italy average over United States average over

Method1970–75

1976–80

1981–85

1986–90

1970–75

1976–80

1981–85

1986–90

Survey of households — — — — 3.7 4.5 5.6 —Tax auditing 3.0 3.9 10.0 4.9 6.3 8.2 10.0Discrepancy bet. expenditure and income 3.2 4.3 9.3 3.2 4.9 6.1 10.2Discrepancy bet. official and actual employment — 18.4 — — — — — —Physical input — — — 19.3 — — 7.8 9.9Currency demand(Tanzi) 11.3 13.2 17.5 21.3 3.5 4.6 5.3 6.2Cash deposit ratio (Gutmann) 23.4 27.2 29.3 — 8.8 11.2 14.6Transactions (Feige) 19.5 26.4 34.3 — 17.3 24.9 21.2 19.4MIMIC (Frey andWeck-Hannemann) — 10.5 — — — 8.2 — —Number of methodsused 5 7 3 4 6 7 7 5

4 Italy: Thomas (1992), Lippert and Walker (1997), Pozo (1996), Schneider (1994a,b, 1997), Bendelac and Clair(1993).5 United States: Thomas (1992), Lippert and Walker (1997), Pozo (1996), Schneider (1994a,b, 1997), Bendelac andClair (1993), Tanzi (1986), Feige (1986), Thomas (1986).


knowledge. First studies on the interac-tion between the shadow economy andcorruption find a positive impact on thelevel of corruption: the larger the cor-ruption, the larger the shadow economy.But clearly, more research is neededhere, too.

Most studies of the shadow economyfocus on the influence on the allocationof resources and the loss of revenue forthe state. But the impact on official in-stitutions, norms, and rules is evenmore important. The shadow economycan be seen as an indicator of a deficit

of legitimacy of the present social orderand the existing rules of official eco-nomic activities. The exit-option shadoweconomy is an important constraint onthe Leviathan state and can help secureeconomic freedom.31

To conclude: we have provided someinformation on the size of the shadoweconomy, and on its causes and conse-quences. But more research is needed

TABLE 9ESTIMATES OF THE SIZE OF THE SHADOW ECONOMY LABOR FORCE IN SOME OECD COUNTRIES

Countries Years

Participantsper 1000people1

Participants as % of

Labor Force2

Size of the ShadowEconomy (as % ofGDP) Currency

Demand Approach3

Sources of Figures forParticipants

Austria 1990–91 300 9.6 5.47 Schneider (1998)1997–98 500 16.0 8.93

Denmark 1980 — 8.3 8.6 Mogensen, Kvist,1986 — 13.0 — Körmendi,1991 — 14.3 11.2 Pedersen1994 — 15.4 17.6 (1995)1998 22.5 18.4 Pedersen (1998)

France 1975–82 800–1500 3.0–6.0 6.9 Raffaele De Grazia1997–98 1400–3200 6.0–12.0 14.7 (1983) and own

calculationsGermany 1974–82 2000–3000 8.8–12.0 10.6 De Grazia (1983)

1997–98 5000 22.0 14.7 Schneider (1998b)Italy 1979

1997–98

4000–7000

6600–11400

20.0–35.0

30.0–48.0

16.7

27.3

D. Gaetani and G. d’Aragona (1979); own calculations

Spain 1979–80 1250–3500 9.6–26.5 19.0 Benito S. M.1997–98 1500–4200 11.5–32.3 23.1 Ruesga (1984);

own calculationsSweden 1978 750 13.0–14.0 13.0 De Grazia (1983)

1997 1150 19.8 19.8 and own calculationsEuropean 1978 10,000 — 14.5 De Grazia (1983) Union 1997–98 20,000 and own calculationsOECD 1978 16,000 — 15.0 De Grazia (1983)

1997–98 35,000 and own calculations

1 Estimated full-time jobs, including unregistered workers, illegal immigrants, and second jobs.2 As percent of the population aged 20–69, survey method. Denmark: as percent of the population aged 20–69,survey method (% heavily engaged in shadow economy activities).3 Source of size of shadow economy: Schneider (1994a, 1998b, 1999).

31 For the importance of institutions and theimpact of the shadow economy, see GeoffreyBrennan and James M. Buchanan (1980, 1985).


to develop a comprehensive, interdisci-plinary, theoretical and empirical ap-proach to learning more about why peo-ple work in the shadow economy andwhat effect it has on the official economy.

REFERENCES

Adam, Markus C. and Victor Ginsburgh. 1985.“The Effects of Irregular Markets on Macro-economic Policy: Some Estimates for Belgium,”Europ. Econ. Rev., 29:1, pp. 15–33.

Adebian E. and B. Standish, eds. 1992. EconomicGrowth in South Africa. Oxford: Oxford U.Press.

Aigner, Dennis; Schneider, Friedrich, and Dama-yanti Ghosh. 1988. “Me and My Shadow: Esti-mating the Size of the US Hidden EconomyFrom Time Series Data,” in Dynamic Econo-metric Modeling. W. A. Barnett, E. R. Berndt,and H. White, eds. Cambridge: Cambridge U.Press, pp. 224–43.

Alessandrini, S. and Bruno Dallago, eds. 1987. TheUnofficial Economy: Consequences and Per-spectives in Different Economic Systems.Gower: Aldershot.

Allingham, Michael G. and Agnar Sandmo. 1972.“Income Tax Evasion: A Theoretical Analysis,”J. Public Econ., 1:3–4, pp. 323–38.

Alm, James. 1996. “Explaining Tax Compliance,”in Exploring the Underground Economy. SusanPozo, ed. Michigan: W.E. Upjohn Institute forEmployment Research, pp. 103–27.

Alm, James; Gary H. McClelland, and William D.Schulze. 1999. “Changing the Social Norm ofTax Compliance by Voting,” Kyklos, 52:2, pp.141–71.

Andreoni, James; Brian Erard, and Jonathan Fein-stein. 1998. “Tax Compliance,” J. Econ. Lit.,36:2, pp. 818–60.

Asea, Patrick K. 1996. “The Informal Sector: Babyor Bath Water?” Carnegie-Rochester Conf. Se-ries Public Policy, 45, pp. 163–71.

Bagachwa, M. S. D. and A. Naho. 1995. “Estimat-ing the Second Economy in Tanzania,” WorldDevelopment, 23:8, pp. 1387–99.

Bagnasco, A., ed. 1986. L’altra Metà Dell’eco-nomia. La Ricerca Internazionale Sull’economiaInformale. Naples: Liguori.

Bardhan, Pranab. 1997. “Corruption and Develop-ment: A Review of Issues,” J. Econ. Lit., 35:3,pp. 1320–46.

Barnett, W. A.; E. R. Berndt, and H. White, eds.1988. Dynamic Econometric Modeling. Cam-bridge: Cambridge U. Press.

Bartlett, Bruce. 1998. Corruption, the Under-ground Economy, and Taxation. WashingtonD.C.: National Center for Policy Analysis, un-published manuscript.

Becker, Gary S. 1965. “A Theory of the Allocationof Time,” Econ. J., 75:299, pp. 493–517.

Bendelac, Jacques and Pierre-Maurice Clair. 1993.

Macroeconomic Measures of the Hidden Econ-omy. International Institute of Public Finance,49th Congress, Berlin.

Bhattacharyya, Dilip K. 1993. “How Does the“Hidden Economy” Affect Consumers’ Expendi-ture? An Econometric Study of the U.K. 1960–1984,” International Institute of Public Fi-nance, Berlin.

———. 1999. “On the Economic Rationale of Es-timating the Hidden Economy,” Econ. J.,109:456, pp. 348–59.

Blades, Derek. 1982. “The Hidden Economy andthe National Accounts,” OECD OccasionalStudies, Paris, pp. 28–44.

Boes, Dieter and Bernhard Felderer, eds. 1989.The Political Economy of Progressive Taxation.Berlin: Springer.

Boeschoten, Werner C. and Marcel M. G. Fase.1984. The Volume of Payments and the InformalEconomy in the Netherlands 1965–1982. M.Nijhoff, Dordrecht.

Bosch, Gerhard and Steffen Lehndorff. 1998. “Ar-beitszeitverkürzung und Beschäftgung: Erfah-rungen in Europa und wirtschaftspolitischeEmpfehlungen,” Vierteljahreshefte zur Wirt-schaftsforschung, 67:4, pp. 300–25.

Brandstaetter, Hermann and Werner Güth, eds.1994. Essays on Economic Psychology. Berlin:Springer:

Brehm, Jack W. 1966. A Theory of PsychologicalReactance. NY: Academic Press.

———. 1972. Responses to Loss of Freedom: ATheory of Psychological Reactance. Morristown:General Learning Press.

Brennan, Geoffrey and James M. Buchanan. 1980.The Power to Tax: Analytical Foundations of aFiscal Constitution. Cambridge: Cambridge U.Press.

———. 1985. The Reason of Rules: ConstitutionalPolitical Economy. Cambridge: Cambridge U.Press.

Cagan, Phillip. 1958. “The Demand for CurrencyRelative to the Total Money Supply,” J. Polit.Econ., 66:3, pp. 302–28.

Cappiello, M.A:. 1986. “Proposita Di BibliografiaRagionata Sull’economia Sommersa Nell’indus-tria, Italia 1970–82,” in L’altra Metà Dell’eco-nomia. La Ricerca Internazionale Sull’economiaInformale. A. Bagnasco, ed. Naples: Liguori,pp. 307–45.

Cassell, Dieter. 1989. “Schattenwirtschaft undDeregulierung: Wohlfahrtsstaatliche Reglemen-tierungen der Wirtschaft als Ursache derExpansion des informellen Sektors und ord-nungspolitische Gestaltungsaufgabe,” in De-regulierung—eine Herausforderung an dieWirtschafts- und Sozialpolitik in der Marktwirt-schaft, Hellmuth St. Seidenfuss, ed. Berlin:Duncker & Humblot Publishing Co., pp. 37–102.

Cassel, Dieter and E. Ulrich Cichy. 1986. “Ex-plaining the Growing Shadow Economy in Eastand West: A Comparative Systems Approach,”Compar. Econ. Stud., 28:1, pp. 200–41.

Cebula, Richard J. 1997. “An Empirical Analysis of


the Impact of Government Tax and AuditingPolicies on the Size of the Underground Econ-omy: The Case of the United States, 1993–94:”Amer. J. Econ. Soc., 56:2, pp. 173–85.

Chickering, Lawrence A. and Muhamed Sa-lahdine, eds. 1991. The Silent Revolution: TheInformal Sector in Five Asian and Near EasternCountries. San Francisco: Int. Center Econ.Growth, ICS Press.

Clotfelter, Charles T. 1983. “Tax Evasion and TaxRates: An Analysis of Individual Return,” Rev.Econ. Statist., 65:3, pp. 363–73.

Contini, Bruno. 1981. “Labor Market Segmenta-tion and the Development of the Parallel Econ-omy—the Italian Experience,” Oxford Econ.Papers, 33:4, pp. 401–12.

———. 1982. “The Second Economy of Italy,” inThe Underground Economy in the United Statesand Abroad. Vito Tanzi, ed. Lexington: D.C.Heath, pp. 131–59.

Cornelius, Peter and Paul Lenain, eds. 1997.Ukraine: Accelerating the Transition to Market.Washington D.C.: IMF.

Cowell, Frank. 1990. Cheating the Government.Cambridge: MIT Press.

Dallago, Bruno. 1985. “Tassazione, EconomiaDell’offerta ed Economia Irregolare,” Eco-nomia, Banca e Congiunta, 1, pp. 27–57.

———. 1990. The Irregular Economy: The “Un-derground Economy” and the “Black LaborMarket,” Dartmouth: U.K. Publishing Co.

De Gijsel, Peter. 1984. “Ökonomische Theorie desSchwarzarbeitsangebots und der Mehrfach-beschäftigung,” in Schattenwirtschaft. Wirt-schafts- und sozialwissenschaftliche Aspekte, in-ternationale Erfahrungen. Klaus Gretschmann,Rolf G. Heinze, and Bernd Mettelsiefen, eds.Göttingen: Vandenhoeck & Rubrecht, pp. 76–96.

De Grazia, Raffaele. 1983. Le Travail Clandestin:Situation Dans Les Pays Industrialisés à Écono-mie de Marché. Geneve: Bit.

De Soto, Hernando. 1989. The Other Path. (ElOtro Sendero. Translated by June Abbott) NY:Harper and Row.

Del Boca, Daniela. 1981. “Parallel Economy andAllocation of Time,” Micros: Quart. J. Micro-economics, 4:2, pp. 13–18.

Del Boca, Daniela and Francesco Forte. 1982.“Recent Empirical Surveys and Theoretical In-terpretations of the Parallel Economy in Italy,”in The Underground Economy in the UnitedStates and Abroad. Vito Tanzi, ed. Lexington:D.C. Heath, pp. 160–78.

Deregulation Commission. 1991. Opening of Mar-kets and Competition, Report Presented to Ger-man Fed. Government, Bonn.

DIW (Deutsches Institut für Wirtschaftsfor-schung). 1998. “Arbeitszeitpräferenzen in West-und Ostdeutschland 1997. Potential für Verkür-zung der Arbeitszeit gesunken,” in DIW-Wo-chenbericht, 65:37, pp. 667–77.

Elster, Jon. 1998. “Emotions and Economic The-ory,” J. Econ. Lit., 36:1, pp. 47–74.

Feige, Edgar L. 1979. “How Big is the IrregularEconomy?” Challenge, 22:1, pp. 5–13.

———. 1986. “A Re-examination of the ‘Under-ground Economy’ in the United States,” IMFStaff Papers, 33:4, pp. 768–81.

———. 1989 ed. The Underground Economies.Tax Evasion and Information Distortion. Cam-bridge: Cambridge U. Press.

———. 1994. “The Underground Economy andthe Currency Enigma,” Supplement to PublicFinance/Finances Publiques, 49, pp. 119–36.

———. 1996. “Overseas Holdings of U.S. Cur-rency and the Underground Economy,” in Ex-ploring the Underground Economy. Susan Pozo,ed. Michigan: W.E. Upjohn Institute for Em-ployment Research, pp. 5–62.

———. 1997. “Revised Estimates of the Under-ground Economy: Implications of U.S. Cur-rency Held Abroad,” in: The UndergroundEconomy: Global Evidences of its Size and Im-pact. Owen Lippert and Michael Walker, eds.Vancouver: Fraser Institute, pp. 151–208.

Feinstein, Jonathan S. 1999. “Approaches for Esti-mating Noncompliance: Examples from FederalTaxation in the United States,” Econ. J.,109:456, pp. 360–69.

Fichtenbaum, Ronald. 1989. “The ProductivitySlowdown and the Underground Economy,”Quart. J. Business Econ., 28:3, pp. 78–90.

Frank, Robert H. 1988. Passion Within Reason.The Strategic Role of the Emotions. NY: Norton.

Franz, Anton. 1983. “Wie groß ist die “schwarze”Wirtschaft?” Mitteilungsblatt der Österreichi-schen Statistischen Gesellschaft, 49:1, pp. 1–6.

Frey, Bruno S. 1989. “How Large (or Small)Should the Underground Economy Be?” in: TheUnderground Economies: Tax Evasion and In-formation Distortion. Edgar L. Feige, ed. NY:Cambridge U. Press, pp. 133–49.

———. 1997. Not Just for the Money. An Eco-nomic Theory of Personal Motivation. Chelten-ham: Elgar Publications.

Frey, Bruno S. and Reiner Eichenberger. 1996.“FOCJ: Competitive Governments for Europe,”Int. Rev. Law Econ., 16, pp. 315–27.

Frey, Bruno S. and Hannelore Weck. 1983a. “Bu-reaucracy and the Shadow Economy: A Macro-Approach” in Anatomy of Government Deficien-cies. Horst Hanusch, ed. Berlin: Springer, pp.89–109.

———. 1983b. “Estimating the Shadow Economy:A ‘Naive’ Approach,” Oxford Econ. Papers, 35,pp. 23–44.

Frey, Bruno S. and Hannelore Weck-Hannemann.1984. “The Hidden Economy as an ‘Unob-served’ Variable,” Europ. Econ. Rev., 26:1–2,pp. 33–53.

Frey, Bruno S.; Felix Oberholzer-Gee, and ReinerEichenberger. 1996. “The Old Lady Visits YourBackyard: A Tale of Morals and Markets,” J.Polit. Econ., 104:6, pp. 193–209.

Frey, Bruno S. and Werner Pommerehne. 1984. “TheHidden Economy: State and Prospect for Mea-surement,” Rev. Income Wealth, 30:1, pp. 1–23.


Frey, Luigi. 1972. Il Lavoro a Domicilio in Lom-bardia, Milan: Giunta Regionale Lombarda,Assessorato al Lavoro.

———. 1975. “Il Potenziale di Lavoro in Italia,”Documenti ISVET, No. 50.

———. 1978. “Il Lavoro Nero Nel 1977 in Italia,”Tendenze della Occupazione, No. 6.

———. 1980. “Introduzione All’analisi Econ.a delLavoro Minorile,” Economia del Lavoro, 1–2,pp. 5–16.

Friedman, Eric; Simon Johnson, Daniel Kauf-mann, and Pablo Zoido-Lobatón. 1999. “Dodg-ing the Grabbing Hand: The Determinants ofUnofficial Activity in 69 Countries,” WorldBank Discussion Paper.

Gaertner, Wulf and Alois Wenig, eds. 1985. TheEconomics of the Shadow Economy. Berlin:Springer.

Gaetani, D and G. d’Aragona. 1979. “I Sommersi,”Nord e Sud, 7, pp. 26–46.

Garcia, Gillian. 1978. “The Currency Ratio andthe Subterranean Economy,” Fin. Analysts J.,69:1, pp. 64–66.

Giles, David E. A. 1999a. “Measuring the HiddenEconomy: Implications for Econometric Model-ling,” Econ. J., 109:456, pp. 370–80.

———. 1999b. “Modelling the Hidden Economyin the Tax Gap in New Zealand,” Econ. DeptWorking Paper., U. Victoria, Canada.

Giles, David E.A.; Linsey M. Tedds, and GugsaWerkneh. 1999. “The Canadian Undergroundand Measured Economies,” Econ. Dept. Work-ing Paper, U. Victoria, Canada.

Gretschmann, Klaus; Rolf G. Heinze, and BerndMettelsiefen, eds. 1984. Schattenwirtschaft.Wirtschafts- und sozialwissenschaftliche As-pekte, internationale Erfahrungen. Göttingen:Vandenhoeck & Rubrecht.

Gutmann, Pierre M. 1977. “The SubterraneanEconomy,” Fin. Analysts J., 34:1, pp. 24–27.

Hanusch, Horst, ed. 1983. Anatomy of Govern-ment Deficiencies. Berlin: Springer.

Halpern, L. and C. Wyplosz, eds. 1998. Hungary:Towards a Market Economy. Cambridge, MA:Cambridge U. Press.

Hartzenburg, G. M. and A. Leimann. 1992. “TheInformal Economy and its Growth Potential,” inEconomic Growth in South Africa. E. Adebianand B. Standish, eds. Oxford: Oxford U. Press,pp. 187–214.

Helberger, Claus and Hans Knepel. 1988. “HowBig is the Shadow Economy? A Re-analysis ofthe Unobserved-Variable Approach of B. S.Frey and H. Weck-Hannemann,” Europ. Econ.J., 32, pp. 965–76.

Hill, Roderick and Muhammed Kabir. 1996. “TaxRates, the Tax Mix, and the Growth of the Un-derground Economy in Canada: What Can WeInfer? Canadian Tax J./Revue Fiscale Canadi-enne, 44:6, pp. 1552–83.

Hirschman, Albert O. 1970. Exit, Voice and Loy-alty. Cambridge: Harvard U. Press.

Houston, John F. 1987. “Estimating the Size andImplications of the Underground Economy,”

Working paper 87–9, Fed. Reserve Bank Phila-delphia.

Hunt, Jennifer. 1999. “Has Work-Sharing Workedin Germany?” Quart. J. Econ., 89:1, pp. 117–48.

IRS. 1979. Estimates of Income Unreported on In-dividual Tax Returns. Washington D.C.: IRS,U.S. Treasury. Dept.

———. 1983. Income Tax Compliance Research:Estimates for 1973–81. Washington D.C.: IRS,U.S. Treasury. Dept.

Isachsen, Arne J. and Steinar Strom. 1980. “TheHidden Economy, the Labor Market and TaxEvasion,” Scandinavian J. Econ., 82, pp. 304–11.

———. 1985. “The Size and Growth of the Hid-den Economy in Norway,” Rev. Income Wealth,31:1, pp. 21–38.

Isachsen, Arne J.; Jan Klovland, and SteinarStrom. 1982. “The Hidden Economy in Nor-way,” in The Underground Economy in theUnited States and Abroad. Vito Tanzi, ed. Lex-ington: D.C. Heath, pp. 209–31.

Jain, Arvind K. ed. 1998. Economics of Corrup-tion. Boston: Kluwer Academic Publishers.

Johnson, Simon; Daniel Kaufmann, and AndreiShleifer. 1997. “The Unofficial Economy inTransition,” Brookings Papers Econ. Act., 0:2,pp. 159–221.

Johnson, Simon; Daniel Kaufmann, and PabloZoido-Lobatón. 1998a. “Regulatory Discretionand the Unofficial Economy,” Amer. Econ.Rev., 88:2, pp. 387–92.

———. 1998b. “Corruption, Public Finances andthe Unofficial Economy,” World Bank Discus-sion Paper.

Juster, F. Thomas and Frank P. Stafford. 1991.“The Allocation of Time: Empirical Findings,Behavioral Models, and the Problems of Mea-surement,” J. Econ. Lit., 29:1, pp. 471–522.

Kaufmann, Daniel. 1997. “The Missing Pillar of aGrowth Strategy for Ukraine,” in Ukraine: Ac-celerating the Transition to Market. P. Cor-nelius and P. Lenain, eds. Washington D.C.:IMF.

Kaufmann, Daniel and Jeffrey Sachs. 1998. “De-terminants of Corruption,” unpublished manu-script, Harvard U.

Kaufmann, Daniel and Aleksander Kaliberda.1996. “Integrating the Unofficial Economy intothe Dynamics of Post Socialist Economies: AFramework of Analyses and Evidence,” WorldBank Policy Research Working Paper 1691.

Kirchgaessner, Gebhard. 1983. “Size and Develop-ment of the West German Shadow Economy,1955–1980,” Zeitschrift für die gesamte Staat-swissenschaft, 139:2, pp. 197–214.

———. 1984. “Verfahren zur Erfassung des in derSchattenwirtschaft erarbeiteten Sozialpro-dukts,” Allgemeines Statistisches Archiv, 68:4,pp. 378–405.

Klovland, Jan. 1984. “Tax Evasion and the De-mand for Currency in Norway and Sweden: IsThere a Hidden Relationship?” Scandinavian J.Econ., 86:4, pp. 423–39.


Korff Wilhelm, ed. 1999. Handbuch der Wirt-schaftsethik. Güthersloh: Gütersloher PublisherCo.

Lackó, Mária. 1996. “Hidden Economy in East-European Countries in International Compari-son,” Laxenburg: International Institute forApplied Systems Analysis (IIASA) working pa-per.

———. 1997. Do Power Consumption Data Tellthe Story?. Electricity Intensity and the HiddenEconomy in Post-Socialist Countries Laxen-burg: International Institute for Applied Sys-tems Analysis. IIASA), working paper.

———. 1998. “The Hidden Economies ofVisegrad Countries in International Compari-son: A Household Electricity Approach.” inHungary: Towards a Market Economy. L.Halpern and C. Wyplosz, eds. Cambridge, MA:Cambridge U. Press, p. 128–52.

———. 1999. “Hidden Economy an UnknownQuantity? Comparative Analyses of HiddenEconomies in Transition Countries in 1989–95,”Working paper 9905, Econ. Dept., U. Linz,Austria.

Lamnek, Siegfried and Jens Luedtke, eds. 1999.Der Sozialstaat zwischen Markt und Hedonis-mus. Obladen: Leske und Budrich.

Lamnek, Siegfried; Gaby Olbrich, and WolfgangSchäfer. 1999. Forschungsprojekt “Devianz imSozialstaat.” Supported by Volkswagen Stiftung,Eichstätt.

Langfeldt, Enno. 1984. “The Unobserved Econ-omy in the Federal Republic of Germany,” inThe Unobserved Economy. Edgar L. Feige,ed.. Cambridge: Cambridge U. Press, pp. 236–60.

Lemieux, Thomas; Bernard Fortin, and PierreFréchette. 1994. “The Effect of Taxes on LaborSupply in the Underground Economy,” Amer.Econ. Rev., 84:1, pp. 231–54.

Lewin, Shira B. 1996. “Economics and Psychology:Lessons for Our Own Day from the EarlyTwentieth Century,” J. Econ. Lit., 34:3, pp.1293–323.

Lippert, Owen and Michael Walker, eds. 1997.The Underground Economy: Global Evidencesof its Size and Impact. Vancouver: The FraserInstitute.

Lizzeri, C. 1979. Mezzogiorno in Controluce.Naples: Enel.

Loayza, Norman V. 1996. “The Economics of theInformal Sector: A Simple Model and SomeEmpirical Evidence from Latin America,”Carnegie-Rochester Conf. Series Public Policy,45, pp. 129–62.

Lubell, Herald. 1991. The Informal Sector in the1980’s and 1990’s. Paris: OECD.

MacAfee, Kerrick. 1980. “A Glimpse of the Hid-den Economy in the National Accounts,” Econ.Trends, 136, pp. 81–87.

Madzarevic, Sanja and Davor Mikulic. 1997.“Measuring the Unofficial Economy by the Sys-tem of National Accounts,” working paper, In-stitute of Public Finance, Zagreb.

Mauleon, Ignacio. 1998. “Quantitative Estimationof the Spanish Underground Economy,” discus-sion paper, U. Salamanka, Spain.

Mauro, Paolo. 1995. “Corruption and Growth,”Quart. J. Econ., 110, pp. 681–712.

Mirus, Rolf and Roger S. Smith. 1997. “Canada’sUnderground Economy: Measurement and Im-plications,” in The Underground Economy:Global Evidence of its Size and Impact. OwenLippert and Michael Walker, eds. Vancouver:Fraser Institute.

Mogensen, Gunnar V.; Hans K. Kvist, Eszter Kör-mendi, and Soren Pedersen. 1995. “TheShadow Economy in Denmark 1994: Measure-ment and Results,” Study No. 3, RockwoolFoundation Research Unit, Copenhagen.

Monopolkommission. 1998. Marktöffnung umfas-send verwirklichen: Hauptgutachten 1996/97.Baden-Baden: Nomos.

Morris, B. 1993. “Editorial Statement,” Interna-tional Economic Insides, IV, International Sta-tistical Yearbook, Budapest.

Neck, Reinhard; Markus Hofreither, and Frie-drich Schneider. 1989. “The Consequences ofProgressive Income Taxation for the ShadowEconomy: Some Theoretical Considerations,” inThe Political Economy of Progressive Taxation.Dieter Boes and Bernhard Felderer, eds. Ber-lin: Springer, pp. 149–76.

O’Higgins, Michael. 1989. “Assessing the Under-ground Economy in the United Kingdom,” inThe Underground Economies: Tax Evasion andInformation Distortion. Edgar L. Feige, ed.Cambridge: Cambridge U. Press, pp. 175–95.

O’Neill, David M. 1983. “Growth of the Under-ground Economy 1950–81: Some Evidencefrom the Current Population Survey,” Study forJoint Economic Committee, 98:122, U.S. Gov.Printing Office, Washington, DC.

OECD. 1998. Employment Outlook. June 1998.Paris: OECD Publications.

Olson, Mancur. 1982. The Rise and Decline of Na-tions. Economic Growth, Stagflation, and SocialRigidities. New Haven: Yale U. Press.

Ostrom, Elinor. 1990. Governing the Commons:The Evolution of Institutions for Collective Ac-tion. Cambridge: Cambridge U. Press.

Park, Tea. 1979. “Reconciliation Between PersonalIncome and Taxable Income,” Bureau of Econ.Analysis, Washington DC.

Pedersen, Soren. 1998. “The Shadow Economy inWestern Europe: Measurement and Results forSelected Countries,” Research Unit Study No.5, Rockwool Foundation, Copenhagen.

Pelzmann, Linde. 1988. Wirtschaftspsycholo-gie: Arbeitslosenforschung, Schattenwirtschaft,Steuerpsychologie. Berlin: Springer.

Petersen, Hans-Georg. 1982. “Size of the PublicSector, Economic Growth and the InformalEconomy: Development Trends in the FederalRepublic of Germany,” Rev. Income Wealth,28:2, pp. 191–215.

Pissarides, Christopher and Gerd Weber. 1988.“An Expenditure-based Estimate of Britain’s


Black Economy,” CLE working paper 104, Lon-don.

Pommerehne, Werner W.; Albert Hart, and BrunoS. Frey. 1994. “Tax Morale, Tax Evasion, andthe Choice of Tax Policy Instruments in Differ-ent Political Systems,” Supplement to Public Fi-nance/Finances Publiques, 49, pp. 52–69.

Pommerehne, Werner W. and FriedrichSchneider. 1985. “The Decline of ProductivityGrowth and the Rise of the Shadow Economyin the U.S.,” Working Paper 85:9, U. Aarhus,Denmark.

Pommerehne, Werner and Hannelore Weck-Hannemann. 1992. “Steuerhinterziehung: Einigeromantische, realistische und nicht zuletzt em-pirische Befunde,” Zeitschrift für Wirtschafts-und Sozialwissenschaften, 112:3, pp. 433–66.

Portes, Alejandro. 1996. “The Informal Economy,”in Exploring the Underground Economy. SusanPozo, ed. Michigan: W.E. Upjohn Institute forEmployment Research, pp. 147–65.

Pozo, Susan, ed. 1996. Exploring the UndergroundEconomy: Studies of Illegal and Unreported Ac-tivity. Michigan: W.E. Upjohn Institute forEmployment Research.

Quirk, Peter J. 1996. “Macroeconomic Implica-tions of Money Laundering,” working paper96:66, IMF, Washington, D.C.

Rabin, Matthew. 1998. “Psychology and Econom-ics,” J. Econ. Lit., 36:1, pp. 11–46.

Riebel, Volker. 1983. Die Schwarzarbeit als Pro-blem der Zeitallokation. Frankfurt am Main:Lang.

———. 1984. “Arbeitszeitverkürzung und Schwar-zarbeit: Auswirkungen einer Verkürzung derWochenarbeitszeit auf das individuelle Arbeit-sangebot,” Zeitschrift für Wirtschafts- undSozialwissenschaften, 104:5, pp. 515–38.

Rogoff, Kenneth. 1998. “Blessing or Curse? For-eign and Underground Demand for EuroNotes,” Econ. Policy: The Europ. Forum, 26,pp. 261–304.

Romer, Paul. 1994. “New Goods, Old Theory, andthe Welfare Costs of Trade Restrictions,” J. De-velopment Econ., 43:1, pp. 5–38.

Rose-Ackermann, Susan. 1978. Corruption: AStudy in Political Economy. NY: AcademicPress.

———. 1997. Corruption and Development.Washington D.C.: The World Bank: AnnualBank Conference on Development Econ.

———. 1999. Corruption and Government:Causes, Consequences and Reforms. Cam-bridge: Cambridge U. Press.

Ruesga, Benito S. M. 1984. “Economia Oculta yMercado de Trabajo: Aproximación al CasoEspanol,” Información Comercial Espanolo, pp.55–61.

Schmölders, Günter. 1960. Das Irrationale in deröffentlichen Finanzwirtschaft: Probleme der Fi-nanzpsychologie. Hamburg: Rowohlt.

———. 1975. Einführung in die Geld- und Fi-nanzpsychologie. Darmstadt: WissenschaftlicheBuchgesellschaft.

Schneider, Friedrich. 1986. “Estimating the Sizeof the Danish Shadow Economy Using the Cur-rency Demand Approach: An Attempt,” Scandi-navian J. Econ., 88:4, pp. 643–68.

———. 1994a. “Measuring the Size and Develop-ment of the Shadow Economy: Can the Causesbe Found and the Obstacles be Overcome?”in Hermann Brandstaetter and Werner Güth,eds. Essays on Economic Psychology, Berlin:Springer, pp. 193–212.

———. 1994b. “Can the Shadow Economy be Re-duced through Major Tax Reforms? An Empiri-cal Investigation for Austria,” Supplement toPublic Finance/Finances Publiques, 49, pp.137–52.

———. 1997. “The Shadow Economies of West-ern Europe,” J. Institute Econ. Affairs, 17:3, pp.42–48.

———. 1998a. “Further Empirical Results of theSize of the Shadow Economy of 17 OECDCountries Over Time,” Paper presented at 54thCongress of IIPF, Cordoba, Argentina and dis-cussion paper, Econ. Dept., U. Linz, Austria.

———. 1998b. “Stellt das Anwachsen der Schwar-zarbeit eine wirtschaftspolitische Herausfor-derung dar? Einige Gedanken aus volks-wirtschaftlicher Sicht,” Mitteilungen desInstituts für angewandte Wirtschaftsforschung(IAW), Linz, 98:1, S. 4–13.

———. 1999. “Ist Schwarzarbeit ein Volkssportgeworden? Ein internationaler Vergleich desAusmaßes der Schwarzarbeit von 1970–97,” inDer Sozialstaat zwischen Markt und Hedonis-mus. Siegfried Lamnek and Jens Luedtke, eds.Obladen: Leske und Budrich, pp. 293–318.

Schneider, Friedrich and Dominik H. Enste.2000. Schattenwirtschaft und Schwarzarbeit.Umfang, Ursachen, Wirkungen und wirtschaft-spolitische Empfehlungen. München: Olden-bourg.

Schneider, Friedrich; Markus F. Hofreither, andReinhard Neck. 1989. “The Consequences ofa Changing Shadow Economy for the Offi-cial Economy: Some Empirical Results for Aus-tria,” in The Political Economy of ProgressiveTaxation. Dieter Boes and Bernhard Felderer,eds. Berlin: Springer, pp. 181–211.

Schneider, Friedrich and Reinhard Neck. 1993.“The Development of the Shadow Economy un-der Changing Tax Systems and Structures,” Fi-nanzarchiv N.F, 50:3, pp. 344–69.

Schneider, Friedrich and Günther Pöll. 1999.“Schattenwirtschaft,” in Handbuch der Wirt-schaftsethik. Wilhelm Korff, ed. Güthersloh:Gütersloher Pub.

Shleifer, Andrei and Robert W. Vishny. 1993.“Corruption,” Quart. J. Econ., 108:4, pp. 559–617.

Simon, Chris P. and Ann D. Witte. 1982. Beatingthe System: The Underground Economy. Bos-ton: Urban House.

Smith, James D. 1985. “Market Motives in the In-formal Economy,” in The Economics of theShadow Economy. Wulf Gaertner and AloisWenig, eds. Berlin: Springer, pp. 161–77.


Smith, Philip. 1994. “Assessing the Size of the Un-derground Economy: The Canadian StatisticalPerspectives,” Canadian Econ. Observer, Cat.No. 11–010, 3.16–33, at 3.18.

Spiro, Peter S. 1993. “Evidence of a Post-GST In-crease in the Underground Economy;” Cana-dian Tax J./Revue Fiscale Canadienne, 41:2, pp.247–58.

Streit, Manfred E. 1984. “The Shadow Economy:A Challenge to the Welfare State?” Ordo Year-book, 35, pp. 109–19.

Tanzi, Vito. 1980. “The Underground Economyin the United States: Estimates and Implica-tions,” Banca Nazionale del Lavoro, 135:4, pp.427–53.

———, ed. 1982. The Underground Economy inthe United States and Abroad. Lexington: D.C.Heath.

———. 1982. “A Second (and More Skeptical)Look at the Underground Economy in theUnited States,” in The Underground Economyin the United States and Abroad. Tanzi, ed.Lexington: D.C. Heath, pp. 38–56.

———. 1983. “The Underground Economy in theUnited States: Annual Estimates, 1930–1980,”IMF Staff Papers, 30:2, pp. 283–305.

———. 1986. “The Underground Economy in theUnited States, Reply to Comments by Feige,Thomas, and Zilberfarb,” IMF Staff Papers,33:4, pp. 799–811.

———. 1994. “Corruption, Governmental Activi-ties, and Markets,” IMF Working Paper 99, pp.1–20.

———. 1998. “Corruption Around the World:Causes, Consequences, Scope, and Cures,”IMF Working Paper 63, pp. 1–39.

———. 1999. “Uses and Abuses of Estimates of

the Underground Economy,” Econ. J., 109:456,pp. 338–40.

Tanzi, Vito and Hamid Davoodi. 1997. “Corrup-tion, Public Investment, and Growth,” IMFWorking Paper 139, pp. 1–23.

Thomas, James J. 1986. “The Underground Econ-omy in the United States: A Comment onTanzi,” IMF Staff Papers, 33:4, pp. 782–89.

———. 1992. Informal Economic Activity. LSEHandbooks in Economics. London: HarvesterWheatsheaf.

———. 1999. “Quantifying the Black Economy:‘Measurement without Theory’ Yet Again?”Econ. J., 109:456, pp. 381–89.

Weck, Hannelore. 1983. Schattenwirtschaft: EineMöglichkeit zur Einschränkung der öffentlichenVerwaltung? Eine ökonomische Analyse. Frank-furt/Main: Lang.

Williams, Colin C. and Jan Windebank. 1995.“Black Market Work in the European Commu-nity: Peripheral Work for Peripheral Locali-ties?” Int. J. Urban and Regional Research,19:1, pp. 23–39.

Witte, A. D. 1987. “The Nature and Extent of Un-reported Activity: A Survey Concentrating onRecent US Research,” in The Unofficial Econ-omy: Consequences and Perspectives in Differ-ent Economic Systems. S. Alessandrini andBruno Dallago, eds. Gower: Aldershot.

Yoo, Tiho, and Jin K. Hyun. 1998. “InternationalComparison of the Black Economy: EmpiricalEvidence Using Micro-Level Data,” paper pre-sented at 1998 Congress of Int. Institute PublicFinance, Cordoba, Argentina.

Zilberfarb, Ben-Zion. 1986. “Estimates of the Un-derground Economy in the United States,1930–80,” IMF Staff Papers, 33:4, pp. 790–98.


shadow economies: size, causes, and · pdf fileschneider and enste: shadow economies: size,...

Documents