The Monetary Method to Measure the Size of the Shadow Economy a Critical Examination

of Its Use
Author(s): Hildegart Ahumada, Facundo Alvaredo and Alfredo Canavese
Source: Revue conomique, Vol. 60, No. 5, L'conomie informelle (Sep., 2009), pp. 1069-1078
Published by: Sciences Po University Press
Stable URL: http://www.jstor.org/stable/40495345
Accessed: 14-05-2016 18:50 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Sciences Po University Press is collaborating with JSTOR to digitize, preserve and extend access to Revue
conomique

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
 The Monetary Method to Measure
the Size of the Shadow Economy
A Critical Examination of its Use

Hildegart Ahumada*
Facundo Alvaredo**
Alfredo Canavese***

The monetary method is a widely used approach to measure the size of the
shadow economy. It is based on the hypothesis that cash is used to make tran-
sactions that agents want to keep hidden from official records. This paper () provides
a formal aggregation framework which stylizes the steps usually followed in empirical
applications and makes clear the assumptions required by this method, (i) demon-
strates that the method has been used implicitly assuming that the income-elasticity
of currency demand is one even in those cases in which its econometric estimate is
not one, (iii) shows that when the money demand function used to estimate the size
of the shadow economy includes the lagged dependent variable, the need to assume
a known initial condition reappears, as it was the case in the early monetary method.

LA MTHODE MONTAIRE POUR L'ESTIMATION DE L'CONOMIE

INFORMELLE : UNE TUDE CRITIQUE DE SES USAGES

La mthode dite montaire est une approche communment utilise dans la lit-
trature pour mesurer la taille de l'conomie informelle. Cette approche est cepen-
dant base sur l'hypothse que les agents utilisent de l'argent en liquide dans les
transactions qu'ils souhaitent dissimuler aux autorits fiscales. Notre article (i) four-
nit une procdure formelle qui donne toutes les tapes ncessaires la mise en
uvre empirique de la mthode montaire et qui rend plus claires les hypothses
requises pour cette mthode ; (i) montre que la mthode montaire implique que
l 'lasticit-revenu de la demande de monnaie est gale un alors mme qu'elle a
t utilise dans des applications empiriques dans lesquelles cette lasticit tait
manifestement diffrente de un ; (iii) montre que, lorsque la fonction de demande
de monnaie utilise pour estimer la taille de l'conomie informelle inclut la varia-
ble dpendante retarde, alors il est ncessaire de supposer que les conditions
initiales sont connues comme dans la mthode montaire classique.

JEL Code: H3-K4

* Universidad Torcuata Di Telia, Miones 2 1 77, C 1 428 ATG Buenos Aires (Argentina). Email:
hahumada@utdt.edu
** Paris School of Economics et CONICET. Correspondance: PSE, 48 boulevard Jourdan,
75014 Paris (France). E-mail: alvaredo@pse.ens.fr
*** Universidad Torcuata Di Telia et CONICET. E-mail: alca@utdt.edu
We thank the referees, F. Schneider and T. Breusch for helpful discussions. Marianne Cornu-
Pauchet provided superb comments. This paper has been prepared for the Conference "conomie
Informelle, Travail au Noir" held at the Universit de Marne-la- Valle in September 2007 and is
based on the authors' previous research on the topic. We gratefully acknowledge the assistance of
Alejandro Francetich and Juan Pablo Xandri.

1069

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
Revue conomique

INTRODUCTION

National accounts do not register a whole set of economic transactions. The

size of such transactions and the causes and consequences of their existence are
studied under different names: hidden, unrecorded, underground, parallel, black
or shadow economy. Undeclared, under-declared, non-measured and under-
registered transactions made to avoid the burden of taxes or to circumvent regu-
lations, illegal transactions connected with crime and corruption and legal but
non-market activities are included in the concept of shadow economy.
Economists have been interested in this topic during the last twenty-five
years although the concepts analyzed are not uniform. Such an interest is tho-
roughly documented by a volume of The Economic Journal in 1999, a survey by
Schneider and Enste [2000] and the recent international seminar held at Marne-
la- Valle University (France [2007]) gathering academic experts and govern-
ment officials concerned with this subject. In recent years the issue also attracted
the attention of the press. The reason is straightforward: underestimating the GDP
implies, for example, an overestimation of the Public Deficit/GDP and Debt/GDP
ratios; therefore, if the size of the unrecorded GDP is important, then any fiscal
and monetary policy decision would be based on biased official figures. Addi-
tionally, country unions (notably the European Union) usually determine budget
targets and member contributions with reference to the GDP and, consequently,
any discrepancy in the shadow economy rates across nations is at the center of
the problem.
The very nature of the shadow economy makes its measurement a difficult
task. Furthermore, different estimation methods target different concepts; authors
do not always seem aware of this observation. Consequently, estimation methods
have become an important theoretical and empirical issue. There are four main
methods used to measure the size of the shadow economy:
1 . Methods based on the estimation of discrepancies between registered aggre-
gate income and aggregated implied microeconomic expenditures identified in
"ad hoc " surveys.
2. Methods based on tax and regulatory surveys.
3. Methods based on observed discrepancies between inputs used (usually
labor and electricity) and outputs declared by industrial firms.
4. Methods based on estimated discrepancies between money required to
finance observed GDP and total currency.

The monetary method is included in the fourth group and is the most widely
used in the literature. It has the broadest scope since it covers all money tran-
sactions (not only industrial or consumption transactions), it does not need costly
surveys restricted to a sample and it allows performing yearly, quarterly or even
monthly estimations to consider the evolution of the size of the shadow economy
with easily available data. The method has been strongly criticized on different
grounds. Caridi and Passerini [2001] argue that it wrongly considers the con-
cepts of unreported and unrecorded activities to be equivalent. Thomas [1999]
and Breusch [2005a], [2005b], among others, doubt about its quantitative accu-
racy based on econometric grounds concerning time series properties, structural
breaks and sensitivity to units of measurement. The meaning of the estimates

1070

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
 Hildegart Ahumada, Facundo Alvar edo, Alfredo Canavese

is unclear, leading to the confusion (visibly present in the literature) between

unrecorded GDP and undeclared transactions to tax authorities. These critiques
are totally valid. l
The main purpose of this paper is to point out two methodological pitfalls that
usually appear in almost all published empirical works that apply the monetary
approach to estimate the size of the shadow economy. The paper also suggests
ways to avoid those common pitfalls.
The paper has five sections. In the first section we review the evolution of the
monetary method used to compute the size of the shadow economy. We remark
that early efforts were based on the assumption that there was a period in history
in which no hidden transactions were done (a known "initial condition"). We
devote the second section to build a formal aggregation framework which sty-
lizes the steps followed in the empirical applications of the method. The aggre-
gation framework shows how the use of econometrics allowed estimations to
get rid of the need to know the "initial condition" and makes clear why equality
of velocities of circulation for the registered and hidden sectors is a common
assumption. It is also used to get the requirements that the income-elasticity of
the demand for real money balances should meet to produce coherent estimates
under that assumption. In section three we suggest a way to estimate directly the
size of the shadow economy avoiding the assumption of equality of velocities of
circulation for the registered and hidden sectors. We devote the fourth section to
point out that the use of short-run econometric estimates reintroduce the problem
of the "initial condition" in the calculation of the size of the shadow economy.
Some closing remarks are presented in the last section.

A REVIEW OF THE MONETARY APPROACH EVOLUTION

The monetary approach to measure the size of the shadow economy is based
on the assumption that cash is used to make transactions that agents want to keep
hidden from official records. Transactions made using cash are difficult to trace
because they leave no tracks. Other assets are registered in financial institutions
and their uses are recorded in such a way that transactions made with them can
be easily inspected. If the amount of currency used to make hidden transactions
can be estimated, then this amount could be multiplied by the income-velocity
of money to get a measure of the size of the shadow economy.
The monetary approach was first presented by Gutmann [1977] and Feige
[ 1 979] and it evolved to use econometric tools in estimates made by Tanzi [1982],
[1983], which are based on Cagan [1958]. The technique was then applied to
measure the size of the shadow economy in the us, Italy, Norway, Canada, South
Africa, Tanzania, Mexico, India, Australia, Austria, Belgium, Denmark, France,
Germany, Great Britain, Ireland, Netherlands, New Zealand, Spain, Sweden,
Switzerland, Argentina, etc.2

1. The System of National Accounts makes strong efforts to include hidden transactions in the
official estimates. Consequently, some undeclared transactions are already incorporated in official
GDP figures.
2. Tanzi [1999] gives a skeptical view given the wide diversity oi the results oDtainea.

1071

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
Revue conomique

Gutmann' s method [1977] is based on four key assumptions: (a) high taxes
and government regulations are the main causes of the existence of a shadow
sector, (b) only cash is used to make transactions in the shadow economy, (c) the
ratio of currency to demand deposits, C/D, is only influenced by changes in
taxes and regulations, and (d) there was some point in time in the past when no
shadow economy existed. As the ratio C/D ofthat period should have prevailed
except for changes in the level of taxes and regulations, increases in C/D are
directly linked to the extra currency used in the shadow economy. The method
assumes that the income-velocity of circulation, v, is equal for the registered and
the hidden economies; hence the size of the hidden sector is v times the extra
currency.

Feige's method [1979] uses the standard version of the quantity theory of
money Mv = PT, where M is money including demand deposits. The value of
transactions is PT. Assuming that the ratio of the value of transactions to nomi-
nal income remains constant through time and that it is known for a period in
which there were no hidden transactions then, total nominal income can be esti-
mated for any period. The difference between estimated total nominal income
and observed nominal income is the size of the shadow economy. Feige assumes
that hidden transactions are done using either cash or checks.
Gutmann and Feige inaugurated two traditions in the use of monetary methods
to measure the size of the shadow economy. Both assume that only some mone-
tary aggregates are used to finance hidden transactions but they differ in the way
the money used in the shadow economy is reckoned: Gutmann' s work is based
on a portfolio selection approach while Feige's paper uses a money demand
calculation.

It is important to point out that both Gutmann's and Feige's estimation

methods required assuming that there was a past period in which no hidden
transactions were done: this "initial condition" is necessary to get the size of the
shadow economy.
The work by Tanzi [1982], [1983] and all the papers based on his approach
perform econometric estimates of the monetary aggregates used to finance hidden
transactions. The introduction of econometric tools recognizes that the income-
velocity depends not only on variables that induce economic agents to make
hidden transactions but also on income and the opportunity cost of holding cash.
The estimated equation of the demand for currency is also useful to get the extra
cash held by economic agents to finance hidden transactions without postulating
that there was some point in time in the past when no shadow economy existed.
It is again assumed that the income-velocity of circulation for registered and
hidden transactions is equal, and so the size of the shadow economy is measured
multiplying the extra cash by the econometrically obtained value of v.

AN AGGREGATION FRAMEWORK

In most published papers, the dependent variable of the econometrically esti-

mated equation of the demand for currency is observed real cash holdings. This
variable is the aggregated value of real cash used to finance registered transac-
tions plus real cash used for hidden transactions. Thus, the estimated demand for

1072

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
 Hildegart Ahumada, Facundo Alvar edo, Alfredo Canavese

currency function should result from the aggregation of demands for real cash
used for each type of transactions.
We devote this section to build a formal aggregation framework to get an
aggregated currency demand function in Cagan' s [1958] tradition1 and to derive
the condition that the income-elasticity should meet to obtain equal velocities
for both, hidden and registered, sectors.
A currency demand function in Cagan's [1958] tradition can be expressed
as:

COf = A(l+e,)aYexp(-T/r) (1)

where Co/ denotes observed cash balances at time t, 6, is the variable which
induces agents to make hidden transactions (for example the ratio of taxes or
government expenditure to GDP)2, Yot is a scale variable (for example registered
GDP), it measures the opportunity cost of holding cash (the interest rate or the rate
of inflation); A, a, and 7 are positive parameters. Observed currency, C^, is
equal to total currency, CTi, which includes cash used for recorded transactions,
CR/, plus cash used for hidden transactions, CHP

Cot = C-I7 = CR/ + CHr ()

Observed income, YOf , is the registered (or reported) income, YR which does
not include hidden income, YHP so

YT = YO + YH = YR, + YHr (3)

Since observed currency includes CH, but observed income excludes Ym,
the usual econometric regression of Cot on Yo/ would result in biased and
inconsistent estimates.

The empirical applications based on the monetary approach follow four steps.
First, a demand for currency is estimated as in (I).3 Second, under the assump-
tion that the demands for CR and CH have the same functional form with equal
parameters, 0 is set equal to zero to get an estimate of the amount of cash
demanded under no incentives to hide transactions, CRt,

Cw = Y&exp(-70- i4)
Third, since Cw is known from (4) and CT( is observed currency, Co then
CHi can be obtained by difference,4
C, = CTi-CRr (5)

1 . Cagan' s type functions are not the only alternative. Bhattacharyya [ 1 990] suggests a different
aggregation in which real cash demanded for hidden transactions depends only on the value of these
transactions and consequently, the observed real cash demand function is not a Cagan's type func-
tion. But most published papers use an aggregate Cagan's demand without performing any explicit
aggregation assumption.
2. The proxy used to measure what induces agents to make hidden transactions nas a cruciai
impact on the estimation. This is an important empirical issue that is beyond the aim of the present
analysis.
3. To take into account that the time series are integrated, some worKs consider equation vu
as a long-run relation. Other papers estimate first difference equations, partial adjustment models
or hybrids.
4. Tanzi [1982] uses CT instead of CT in (5).

1073

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
Revue conomique

Fourth, to get the size of the shadow economy it is assumed that the velocity
of circulation for both, registered and hidden transactions, is the same, so
Y Y
v _ 1R/ _ XHf (f.'
^Rt ^Ht

Then,
%, = Vr,CH( (7)
YHi is "the" estimation of the size of the shadow economy and it is obtained using
CHt from (5) and vRi from (6).1
So far we have described the procedures followed in the literature. The key
assumption made explicit in (6) requires = 1, which is evident if we recall that
the income-velocity for registered money is
Y Y Y1-
_ - lRt - _

Vr' _ - Cw - _

while the velocity for the hidden economy is

Y Y Y1-
LU __

V"' CH( __

The velocity is the same in both sectors if = I.2 Those studies that find
t 1 but follow the steps described above are therefore incorrect.3
Equations (8) and (9) show the precise relationship between vH and vR since

?*=(M~- VRt ' 1Rt I (io)

VRt ' 1Rt I

Equation (10) demonstrates that when the assumption of equal velocities is

replaced by the assumption of equal elasticities, the empirically expected result
that vH < vR is obtained for the usual case in which < 1 and YH < YR .
It is important to point out that the assumption that the demands for CR and
CH follow Cagan's type forms with equal parameters (A, and 7) is sufficient
for an explicit aggregation of (2) as

CT, = AYexp(-7i,) + AYexp(-Ti,) = AYexp(-7i,) 1 +(^) J. (11)

The usual practice implicitly estimates the Cagan's function (1 1) without dis-
cussing the aggregation issue. A possibility for aggregation is to impose equality
of A, and 7 in both sectors. Even in this special case in which the assumption
of equal parameters for both demand functions is accepted, equality of velocities
requires = 1.

1. It should be clear here that the question about which concept of the shadow economy (unre-
corded GDP, undeclared transactions, etc.) is being measured by expression (7) remains unresolved.
2. The velocity is also the same for the improbable case in which YR = YH for any .
3. It should be stressed that currency is the money aggregate whose demand should have an
income-elasticity equal to one. While this value may appear reasonable and theoretically-based in
the case of the demand for the aggregate used to finance all transactions (e.g. the demand for Ml), it
may not necessarily be correct for narrower definitions of money. For instance, in Baumol-Tobin's
model, the value of transactions elasticity is 1/2.

1074

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
 Hildegart Ahumada, Facundo Alvar edo, Alfredo Canavese

This formulation is also valid for any wider aggregate (e.g. Ml ) as long as
the interest rate or the inflation rate is its opportunity cost. However obtained
Y
from (1) only matches in (1 1) if the ratio -^ is independent of YRt . 1 It must
R/
be stressed that all papers using this approach implicitly make this assumption.

A DIRECT ESTIMATION

In this section we show that the four steps usually followed to compute the
size of the shadow economy (as described in the section "An Aggregation
Framework") can be replaced by a direct estimation explicitly recognizing the
dependence of v on the value of .
Equation (1 1) can be written as:
CT, = AYexp(-O(l+e()a (12)
which is equal to (1).
Recalling that Yo = YR, CT = Co and that CT and YR are observed variables,
(12) can be econometrically estimated as in (I).2 Next, as we already described
in the previous section, setting 0, = 0 provides an estimate for CH. The ratio
between CR and CH is

CR, AY|,exp(-TQ /YR,f

Cw~AYexp(-7i,) KJ'
Equation (13) provides an expression for YHi given YR/,CR,,CH, and .3
We do not claim that this method provides the correct way to estimate the velo-
cities of circulation both in the formal and informal sectors. We only argue that the
technique itself is not only based on the idea that the velocity is lower in the shadow
economy (in which "everything" is financed with cash) but it also provides two
distinct estimates, so that the assumption of equality is wrong and unnecessary.
It should be mentioned that when the ratio of two monetary aggregates is
econometrically estimated instead of a money demand equation, the same
objection discussed above remains unsolved. Once money used to finance non-
registered transactions is calculated, the assumption of equal velocities of cir-
culation for registered and hidden currencies is again made in the literature to
get the size of the shadow economy. In fact, velocity should be "known" to get
transactions. The critical point is that velocity depends on the income elasticity
of money demand which is an empirical issue. Indeed, this case is even worse
since no estimations of such elasticity are provided by modeling ratios.4

1 . For this reason it is convenient to measure 0 normalized by registered GDP.

2. The variable (l + 0)a is sensitive to changes in the units in which fc) is measured, as pointed
out in Breusch [2005a], [2005b]. We keep the notation to maintain Tanzi's original functional form.
Nevertheless, changing (l + 0)a for (exp(a0) - l) could solve this problem.
3. Ahumada et al. [2007] suggest a way to correct the estimated size o the shadow economy
when the income elasticity of the demand for money is not one and apply the correction to existent
measures for different countries.
4. Itshouldbenoticedthatasignificanteffectottransactions(asin lanzi s econometric equation;
indicates different elasticities for the aggregates entering the ratio and thus different velocities.

1075

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
Revue conomique

SHORT-RUN VS. LONG RUN ESTIMATES

The way suggested in the section "A Direct Estimation" to get the size of the
shadow economy is based on an aggregate Cagan's type money demand with no
dynamic structure. During the last twenty years more accurate representations
of the money demand have been developed to take into account the time series
properties of the data involved (cash holdings, income, interest rates, inflation),
which are either integrated or highly persistent. From "partial adjustment" to
"equilibrium correction", these dynamic models admit two types of estimates:
one for the short-run and another for the long-run. Thus, long-run estimates
(as those performed using equation 1) were replaced by short-run estimates
as a central input to the method. When the short-run models that include the
lagged dependent variable are used to obtain the size of the shadow economy,
the problem of the "initial condition" reappears: previous values of the size of
the shadow economy (and eventually a period without hidden transactions) must
be identified to get a measurement. No need of such "knowledge" appeared to
be a remarkable advantage of the econometric approaches, when compared with
previous methods.
To analyze this issue, we consider a usually estimated log form of regression
that includes the lagged dependent variable,
cTt = bx + b2cTt_ x + b3yrt + b4it + 5ln(l + O,) (14)
where lower case letters stand for logarithms. Equation (14) can be seen as a
model of partial adjustment to reach

C^AY&expHOO+,)01
which is the same as equation (12), or in logs:
cTt = ln(A) + $yRt - vt + aln(l + 6,) (15)
as the equilibrium value for the desired level of real cash holdings. Equation (14)
has the following long-run solution
*

^+t^t^tV1'^- (16)
It can be readily seen from (15) and (16) that:

b' - ln(A). h _. b4 _ r b5
In this case it is straightforward to show that when 0, = 0, CRt can be directly
obtained. However to get CRi for the short run from equation (14) is a quite dif-
ferent issue since C^. l5 or a previous value of it, should be known. To further
analyze this issue it is convenient to show the aggregation assumed for CT in the
partial adjustment model. To get equation (14) taking logs1 it should be

CTi = C^_,C-M = (CR_1 + Ch,.,)^ + C*HI)(1-W (17)

1. If partial adjustments were assumed for each component (CR and CH) in equation (17), this
Y Y C
equation would not allow us to estimate the share ~, as it cannot be isolated from R R/-1 unless
W/-1 - ^H-l-

1076

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
 Hildegart Ahumada, Facundo Alvar edo, Alfredo Canavese

where

C; = AY exp(-Ti,) 1 + (^) = AY exp(-7/,)(l + f)a. (18)

When Qt = O, only that part of CR (target or desired) due to registered tran-
sactions can be obtained but not CR/ as a whole since CRt_l should also be
known. In most applied work CTt _ x is used instead of CRi _ 1? and so CRt is over-
estimated as estimates are positive. If CR,_ x were substituted by CRi_2 and
CRi_2 by its previous value and so on, the required knowledge about registered
currency would be moved back to an initial value CR0 (the initial condition of
the solution of the pertinent first order difference equation for the log of currency
demand). But CR0 is registered real cash for t = 0 which can only be identified
if there are no hidden transactions so that CR0 = CT0.
To sum up, the only way to avoid ad hoc assumptions about previous values of
registered currency is to restrict the measures of the size of the shadow economy
to those based on the long-run estimates of the money demand.1

FINAL REMARKS

The monetary method to measure the size of the shadow economy is based
on econometric estimates of the demand for money. These estimates are used to
get the currency held by economic agents in excess of the amount they need to
finance registered transactions. The standard monetary approach uses the excess
of currency multiplied by the velocity of circulation (assumed to be equal in the
registered and the shadow economies) to measure hidden income.
In this paper we analyze two methodological pitfalls that are present in most
empirical applications of the monetary approach.
The first pitfall appears when it is assumed that velocities of circulation are
equal for registered and hidden sectors. This assumption is accurate only when
the income elasticity of the demand for money is one.
The second pitfall appears when short-run representations of the money
demand that include the lagged dependent variable are used to estimate the size
of the shadow economy. In this case, a past value of the demand for cash used
for hidden transactions should be known and this implies assuming that there is
a point in time in which no shadow economy existed. So, the only way to avoid
ad hoc assumptions about previous values of registered currency is to restrict the
measures of the shadow economy size to those based on the long run estimates
of the money demand.

1 . An alternative presentation of this issue appears in Ahumada et al. [2008].

1077

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms
Revue conomique

REFERENCES

Ahumada, H., Alvaredo, F. and Canavese, A. [2007], "The Monetary Method and
the Size of the Shadow Economy: A Critical Assessment," Review of Income and
Wealth, 53 (2): 363-371.
Ahumada, H., Alvaredo, F. and Canavese, A. [2008], "The Monetary Method to
Measure the Shadow Economy: The Forgotten Problem of the Initial Conditions,"
Economics Letters, 101 (2): 97-99.
BHATTACHARYYA, D. [1990], "An Econometric Method of Estimating the Hidden
Economy, United Kingdom (1960-1984): Estimates and Tests," Economic Journal,
100 (402): 703-717.
Breusch, T. [2005a], "Australia's Cash Economy: Are the estimates credible?," The
Australian National University, Mimeo.
Breusch, T. [2005b], "Fragility of Tanzi's Method of Estimating the Underground
Economy," The Australian National University, Mimeo.
Cagan, P. [1958], "The Demand for Currency Relative to the Total Money Supply,"
Journal of Political Economy, 66 (4): 303-328.
Caridi, P. and Passerini, P. [2001], "The Underground Economy, the Demand for
Currency Approach and the Analysis of Discrepancies: Some Recent European
Experience," Review of Income and Wealth, 47 (2): 239-250.
Feige, E. [19791, "How Big is the Irregular Economy?," Challenge, 22 (1): 5-13.
GUTMANN, P. [1977], "The Subterranean Economy ," Financial Analysts Journal, 34 (1):
24-27.
Schneider, F. and Enste, D. [2000], "Shadow Economies: Size, Causes, and Conse-
quences," Journal of Economic Literature, 38 (1): 73-1 10.
Tanzi, V. [1982], "Underground Economy and Tax Evasion in the United States: Esti-
mates and Implications," in V. Tanzi (ed.), The Underground Economy in the United
States and Abroad, Lexington Books.
Tanzi, V. [1983], "The Underground Economy in the United States: Annual Estimates,
1930-80 " IMF Staff Papers, 30 (2): 799-81 1.
Tanzi, V. [1999], "Uses and Abuses of Estimates of the Underground Economy," Eco-
nomic Journal, 109 (456): 338-347.
Thomas, J. [1999], "Quantifying the Black Economy: 'Measurement without Theory'
yet again? ," Economic Journal, 109: 381-389.

1078

Revue conomique - vol. 60, N 5, septembre 2009, p. 1069-1078

This content downloaded from 146.102.116.51 on Sat, 14 May 2016 18:50:36 UTC
All use subject to http://about.jstor.org/terms