The Economy of Narco-Dollars: From Production to Recycling of Earnings

For an economist, the problems presented by the production, sale, and, use of drugs both reveal the limits of his discipline and strongly encourage the study of such problems, The object of study is poorly defined, its measurement is to say the least, difficult and often 'folkloric'², the behavior of the dealers is relatively unknown, and their possible change in status is difficult to determine.
 

The object of study is poorly defined precisely because its definition depends on a prohibition that varies according to the country and period of time. The consumption of coca leaf is authorized in certain countries, but forbidden in most. Trafficking is prohibited in certain countries, while the use of drugs is not repressed in others. The amount of variation is considerable and little is known about its modalities: the differentiation can be horizontal or vertical, according to the kind of products and above all to the degree of purity, which itself varies according to the forms of repression and the fluctuation in prices. Quality therefore is difficult to assess, the variation not being defined before the act of sale by the dealers. Likewise, little is known about substitution in products, which depends upon the differentiated change in prices, the amount of dependency, and the changes in the cultural context. The rapid rise of synthetic products-new chemical cocktails-is significant; their use has substituted in part for the use of natural drugs, which are made from plants converted by the help of chemicals, and they are often mixed with these natural forms. The distinction between what is medicinal(thus legal because administered by prescription) and what is not, is not always an easy one, especially if these substances help increase performance, such as speed and endurance. The professionalization of sports and the extraordinary commodification of this domain leads naturally to the "doping" of athletes, Drugs become, therefore, a part of the reproduction of the workforce of athletes. The strong inflow of these substances is revealing of profound societal problerns,^[3] but also of the difficulties that arise in defining what is a drug and what is a medicine ^[4] and of the limits and often arbitrariness of the law. These represent old problems since they were encountered many times during the international discussions concerning legalization at the end of the last century ( concerning opium primarily) and at the beginning of this one.^[5] But there are also new problems in this debate, since it is now a question of synthetic substances the short- and long-terms effects of which are difficult to determine on the health of those who try multiple cocktails composed of more or less mysterious substances.
 

Measurement is imperfect mainly because it is a question of substances whose production, conversion, and commercialization are illegal and whose estimates, as we will see, are often "folkloric," These estimates are all the more arduous to make, since the organizational structures of commercialization, in its different stages, are formed within the networks of informal activities that act as their support, and they themselves take on the aspect of changing and diverse networks-far from the image given by the press when it evokes this or that cartel. Paradoxically, we can obtain an estimation, or more precisely a macro-economic range, which is credible in regard to the production of drugs and its value, On the other hand, the estimation of sums that are repatriated and directly attributable to the se criminal activities is more problematic.
 

The behavior of drug dealers is difficult to define and to judge. The increasing opening up of economies, as much at the level of the ex change of merchandise as at that of the movement of capital, facilitates the exportation of illegal substances and makes the laundering of money apparently easier. Paradoxically, however, it also raises their cost, as we shall see. The beginning of deep crises in numerous ex-socialist economies "in transition towards capitalism" (what used to be called "emergent economies"), the maintenance of quasi-autocracies of certain Asiatic regions (be it in countries like Burma or regions grouping several countries together), apart from the nature of the illicit commerce itself, tend to multiply the supply precisely when the demand in some of the most developed and important countries tends either to stagnate, to regress, or to diversify towards more synthetic substances, and when the efficiency of repression seems to increase at the level of confiscation, This behavior is even more difficult to assess when it is a question of determining the amount of money repatriated into the countries of production, At what level in the chain of commercialization (wholesale, semi-wholesale, retail) does this behavior cease? This is a thorny question, since we know there are multiple factors that increase the particularly high prices between the cost at the production, wholesale at embarkation, semi-wholesale and retail levels.^[6] At precisely which point in the equation does arbitrariness begin to enter, when we posit a hypothesis that the prices upon which we will base our estimation of possible repatriation are the wholesale prices at arrival for cocaine, whereas we rely on the wholesale prices at departure for heroin, for Colombian drug dealers? Finally, beyond this question, what constitutes this repatriation?
 

The techniques of laundering, as sophisticated as they are, cannot circumvent one essential question, that of the status of this money. What legitimizes the possession of large sums of clean money? The response to this question is fundamental and outlines the limits of the search for "notability" by drug dealers, To the extent that it seems easier in many countries with lax legislation to "legitimate clean money" when it is used in construction, building speculation, or land purchase, we understand the preference the drug dealers have for these activities and these countries-but also the difficulties they have in transforming themselves into "bourgeois industrialists."
 

The first purpose of this article is to outline the problems raised by the commercialization of illicit "natural substances," Secondly, it is to present the different techniques that permit the repatriation of dirty money and its laundering, and then to assess the importance of these acts of repatriation of funds from a macro-economic point of view. Lastly, the article proposes to analyze the behavior of the Mafia entrepreneurs.

ASSESSMENTS OF THE GLOBAL PRODUCTION AND CONSUMPTION OF COCAINE

The least we can say is that the estimation of drugs produced and consumed is arrived at through imperfect information. The observer does not have reliable facts; and the producer, the dealer, and the consumer are each ignorant, but to varying degrees, of the macro-economic facts of the market, The use of probabilities is difficult, since only the ranges of price and production can have a degree of satisfactory credibility. To our knowledge, not much has come from the use of game theory in these matters;^[7] and the techniques of industrial economics, which attempt to define behavior through imperfect information-like that of unknown moral factors or adverse selection-help little today, so completely is the information imperfect and its verification a posteriori difficult to compare in order to re-evaluate this behavior. The most credible approach probably consists of combining the facts and the estimates obtained before (production and conversion) and after (consumption), This is the approach that we will adopt, as it is the only one that makes the results obtained on the side of supply coherent.
 

The "upstream"-the beginnings of the process-analysis consists of making a series of assessments. Let us consider the case of cocaine, probably the most studied in the literature, We can determine the number of hectares devoted to the cultivation of coca leaf by selecting the countries likely to produce it (mainly the Andean countries: Peru, Bolivia, Colombia, but also Ecuador, to which one should probably add other countries whose supply, however, until now has appeared relatively marginal), We then determine the amount produced per hectare, which varies according to the fertility of the land, the fertilizer used, and climatic changes.^[8] We obtain a range of quantities produced, from which one should subtract the local consumption of coca leaves, which is significant in Peru and Bolivia. Once this is deducted, we obtain a quantity of leaves whose conversion into "paste," then into "base," constitutes the relatively simple stages of the conversion process, This transformation takes place by the addition of various chemicals in laboratories and results in cocaine hydrochloride, which is cocaine, Table 1 gives an approximation of the production of leaves in each of the producing Andean countries. To better evaluate the estimate of the conversion from leaves to cocaine, we assumed that they were converted in their country of origin at the rate of only 80% in order to take into account local consumption. These factors of conversion, different according to the quality of the leaves, are estimated for Peru at 334 (thousands of tons)/1 (ton); Bolivia at 373/1; and Colombia at 500/1.

Table 1. The Production of Coca Leaves (in Thousands of Tons) and of Cocaine (in Tons)
 

Source: (Steiner 1997: 27). NB. Coca leaves (potential production) in thousands of tons and cocaine in tons.

The simplified hypothesis of location identity concerning the cultivation of leaves and their transformation into cocaine, which is necessary to account for the different factors of conversion, does not correspond to reality. The transformation is not in fact localized to the site of production. One country dominates all of the others by far: Colombia. The criminal organizations in Colombia import base-cocaine from Bolivia and Peru, which added to that produced in Colombia, is changed into cocaine and then exported primarily to the United States, The division of labor between the countries that produce the raw materials without changing them into cocaine and the country that effects this transformation tends, however, to vary. We assume for example that the participation of Bolivia increased the se past years, since it converted a little more than one third of the base-cocaine in 1990, even though this number was only 7% in 1986; at the same time it increased its own production of base-cocaine^[9] significantly and developed its exports in Brazil.^[10] We estimate that in 1990, Bolivia exported 14 tons of base cocaine and 61 tons of cocaine; Peru 350 and 40 tons respectively; and Colombia, approximately 70% of the cocaine produced in the world, 455 tons.
 

To understand the value of the cocaine exported, one must multiply the net quantity ^[11] produced by a price or an average for a range of prices, Different prices should be taken into consideration: the wholesale price at departure, the arrival price in the consuming countries, the semi-wholesale price, and the retail price. The supposition is that Colombia controls the transportation and that it is necessary, therefore, to consider the wholesale prices at arrival in order to deduct the sum of money that could be repatriated once laundered. This assumption is maintained for two reasons: the first is that part of the criminal activities in the countries of destination is also the result of Colombian networks and that, consequently, their participation in the chain from production to final consumption is not limited to transformation and transportation, The second is that a significant part of the transportation derives from the growing participation and in-creasing importance of Mexican criminal networks,^[12] simultaneously with changes in routes. Whatever the case may be, once this assumption is accepted, the possible repatriation can be calculated year after year, so that its estimate intersects with the estimates of those activities that make it possible (contraband, over- and under-billing, and so on), which we will analyze later, taking into account the variations of supply, which are very high and clearly oriented toward a decline in wholesale prices-which ranged from a little more than 50,000 dollars a kilo on average in 1981 to slightly above 10,000 dollars in 1994, after reaching a trough in 1991 (Rocha in Thoumi 1998: 155).^[13]
 

The figures obtained, once the international confiscations are deducted, are only credible provided that the estimates made of the supply are close to those carried out in respect to demand. The credibility of the estimate rests, therefore, on the comparison between the estimates of production and consumption. What remains, then, is to assess the amount of consumption. A simple but deceiving way to determine the estimation consists of multiplying the quantities confiscated by ten-those quantities, that is, that are known, This approach is, however, not very credible: consumption would appear very high and significantly above the maximum estimates of production, Another way to proceed is to distinguish, through research, the occasion al consumers from those that have become dependent.
 

Once the total cost is known and divided by an average retail price range, we obtain the estimation of consumption in volume that we can therefore compare to the estimations in regard to supply. In this way, we obtain for the United States an estimate of consumption of 244 tons (minimum estimate) to 311 tons (maximum estimate) in 1988. These figures are much lower than the estimates made by The Economist in 1989, which were taken from the works of a sub-committee of the U .S. Senate, in which the global traffic of drugs was estimated at about 500 billion dollars, 300 million of which were for the United States alone, one-third of which was for cocaine. Our estimations are also much lower than the estimate of 100 billion dollars, which Steiner qualified as "folkloric" (1997: 6,23) and which was presented by Fortune Magazine, without a description of the methodology used to obtain such a number. This estimation is not only often evoked in the press, but also presented by researchers and in serious studies, where there is little concern given to the macro-economic implications of such a figure (IMF, De Maillard: 1998).^[14] Divided by wholesale prices that prevailed in this period ( about 40,000 dollars per kilo ), consumption would have been about 2,500 tons, and when divided by retail prices, somewhat more than 800 tons (!). Be that as it may, after a peak in 1989, consumption decreased and then settled at between 224 tons and 283 tons in 1993.
 

When we take into account the consumption of other countries and we add the confiscations, we obtain an estimate of world exportations of 571 tons on average from 1989 to 1993 (see Table 2). If we consider that Colombian exports correspond to 75% of global exports, which is more reliable than the slightly lower estimate of De Rementeria (see below), we obtain the amount of exports of this country, that is to say, production net of local consumption. If this estimate corresponds to that obtained from the analysis of supply, we consider it to be globally pertinent. This is in fact the case. The global average of cocaine consumption, from 1988 to 1993, is situated at about 294 tons and the confiscations at about 277 tons, The total exports are thus around 571 tons on average for the time period, The estimated average production, for the same time period is 628 tons according to Steiner (see Table 1 ). The difference between the two estimates is thus approximately 10%, which is low and in certain years very low (for ex ample in 1989), but high in other years (above all in 1990). This difference would be on average less if we had taken the maximum estimate of consumption and not the average between the two estimates. The two estimations-production and consumption-thus appear credible, because there is coherence between them.
 

Table 2. Net Consumption, Confiscation. and Exportation of Cocaine

(1) Assuming that the United States represents 90% of the global market.
(2) Assuming that Colombia supplies 75% of the total.
Source. (Steiner 1997. 24). The production estimates are based on the calculations of the author. The consumption estimates are based on information provided by the White House Office of National Drug Control Policy.

Two conclusions can be deduced from these figures. The first is that cocaine consumption tends to decrease in the United States. at times when the price decreases greatly, The estimate of the number of cocaine deals, be it at the wholesale or retail level, is well below those that we generally find in the press. The second is that confiscations are at a very high level, significantly above the estimates widely made, since they would be located on average at 90% of global consumption, which is a little less than 50% of global production, To downgrade the importance of confiscations is to render incoherent the intersection of established facts on the side of supply and demand and either to over-estimate consumption or to under-estimate production, or both. We are far from the "folkloric" estimates announced here and there-very often by the official organizations-whose objective seems more to be the war against crime rather than the scientific presentation of the drugs economy.
 
 
 

REPATRIATING-LAUNDERING IN COLOMBIA

To determine estimates concerning repatriation of narco-dollars is difficult for two reasons: the first concerns the motivations for repatriating capital; the second concerns laundering itself and the different techniques used. The analysis that we are going to present will be centered on Colombia-the principal global producer of cocaine, a recent producer of heroin, and a base for the illegal exportation of these substances, in addition to marijuana and emeralds. The reference to the totality of illegal trade in drugs and emeralds is to point up the difficulty in attributing to this or that criminal activity the origin of transfers of dirty money.
 

The motivations for repatriating are difficult to define. Why would a Colombian criminal organization be interested in repatriating capital from the United States to Colombia? It could very well leave a substantial part of its earnings in American banks, or in others, once laundered. To evoke the nationalism of Colombian mafias is an insufficient argument, even though it should probably be taken into consideration, as should cultural orientations of the killers tied to this trafficking, who are strongly influenced by religion, cross themselves before committing their deeds, and who thank God for the success of their operations, Another argument appears more pertinent: laundering is more than an ensemble of techniques aimed at transforming, or in other words, changing the form of "dirty" money. It also involves a change of "phase," according to an expression of a Cali Cartel (F. Jurado ), borrowed by De Maillard, that is to say, it gives the money a status and it in this way, makes it honorable. In other words, it does not suffice simply to launder dirty money: acquisition of capital made "clean" in this way must have a plausible justification. Here resides the main difficulty. Geographic proximity can be thought of as reducing transaction costs and making it easier to give a clean status to repatriated capital. The search for this change in status would, therefore, explain in part the repatriating of funds. We will see in what follows that it does not suffice to give "noble titles" to the mafias, that the "nobilization" of the latter is difficult and, as such, makes their transformation into ordinary entrepreneurs in one generation an arbitrary process, In any case, the search to give an honorable status to laundered and repatriated money influences the choice of the techniques used for laundering. While laundering-repatriating does not always succeed in giving an honorable status to money, it does however, follow set paths: it is invested in real estate, cattle raising, and financial speculation. Besides the advantages offered by geography-a country characterized by a large informal sector, ease in circumventing the law and expansive corruption-these investments can be called recycling-laundering in offering a status to laundered money, In this way, laundering also further aids more laundering.
 

The object of this section is not to explore at length the multiple ways to repatriate and launder dirty money. This has been done elsewhere, and in general, it has been done well (reports by The Financial Action Task Force on Money Laundering; Kopp: 1995; De Maillard: 1998: Dupuy: 1998; Thoumi: 1997; Geffray: 1996 and 1998). The object here is probably less technical and more inductive, since it aims to show that these techniques impose particular patterns of behavior. This behavior eventually makes the nobilization of certain mafias difficult, limits their areas of investment to activities that support laundering (hotels, restaurants, gambling halls, and so on) and speculative ventures (cattle raising, real estate construction, highly-rated stocks), and ultimately determines the form taken by their prestige.
 

The techniques used are numerous and evolve with time according to the changes in national and international regulations. The particularity of laundering in this case is that it includes the conversion of one currency into another, here the dollar, the strong currency, against a local currency, the weak currency. This is why a distinction needs to be made between what we call repatriating-Laundering and recycling-laundering. The two movements certainly intersect and feed one another, but the problems raised as a result of each are different.
 

The simplest techniques for repatriating-laundering consist of, first, the mailing of 100-dollar bills by Colombians residing in the United States to their families, secondly, the wiring/transferring of the maximum amount of funds authorized by current legislation,^[15] and finally, the use of "mules" who transport dollars on the return trip to Colombia after having "swallowed" bags of cocaine on their trip to the United States. The sums transferred or transported in these ways are considerable, although modest in view of the size of the sums available for transfer. These techniques remain, however, artisanal.^{[16 ]}When ex change controls exist, which was the case not very long ago, the technique of clearing can be used, This consists of furnishing currencies to a non-resident who wants to visit the United States as a tourist, in ex change for compensation in a Latin American country, Clearing can also be used when the desire of industrialists to place capital illegally outside of their country coincides with that of criminal organizations who want to repatriate a part of their profits. In this case, because of the size of the sums in question, a preliminary laundering in the United States is necessary. These techniques can be sophisticated, while still remaining artisanal, when we take into account the rate of exchange-official and unofficial-the interest rates-both domestic and foreign-and their respective changes (this is, moreover, why we can in part estimate the size of these movements by changes in the differential rate [Urrutia and Ponton 1993]).
 

There exist three major ways to repatriate-launder dirty money: contra-band, over- and under-billing of exported and imported merchandise, and the use of international financial markets.
 

The under-billing of imports is interesting to analyze because it puts into play several factors. On the one hand, in order to be effective, it requires the establishing of a vast network of complicity, since it deals with the manipulation of prices and, therefore, businesses in order to launder the dirty money, On the other hand, it causes a classic arbitrage to intervene between the different rates of ex change, Let us give anexample: during a period of regulated ex change, we observe in general the coexistence of two rates of ex change, one official and the other unofficial. The size of the funds transferred that were due to the criminal activities we are considering led to a paradoxical situation in Colombia: the unofficial rate of ex change was estimated on the basis of the official rate of ex change for long periods of time. The transferring of funds became therefore relatively less profitable than the technique of under-billing, since it was practiced at the official rat y of exchange.^[17] Conversely, under-billing allowed for greater acquisition by the "washed" dollar than by local currencies. Finally, even though we can only present them in the context of this article, the estimates presented here were made by joining together the changes in the differential rate of interest and under-billing (Steiner 1997: 72 ff.] ).
 

The sums transferred through manipulations of prices were considerable (see Table 3). Their determination, although obviously approximate, is nevertheless quite reliable. It consists of comparing the registered prices of the businesses who export to Colombia and the announced prices at the level of imports, and of a correction of this difference by a coefficient of rectification which takes into account the FOE export price and the CIF import price, as well as delays.

Table 3: Under-Billing ( - ), Over-Billing ( + ), Colombian Imports According to Different Authors (in Millions of Dollars)
 

Source. (CID 1997: 28). The methodology of the CID is the same as that of Steiner and Rodriguez.

Although the sums transferred fluctuate and often indicate different estimates depending on the author, at times they reach very high levels ( with a peak of more than 1, 7 billion dollars in 1992 in Colombia ( CID 1997: 28). It is sufficient for the rates of ex change and the interest rates to operate differently in order for over-billing to replace under-billing (in 1993 and 1994) as the means of laundering "narco-dollars," even though the size of the sums transferred by this method is more modest and its occurrence is rarer.
 

It may appear paradoxical that contraband should still play an important role at a time when borders are opening up due to the liberalization of economies that began about ten years ago. One can certainly attempt to explain this by the differentials in the rate of indirect taxation, notably for alcohol and cigarettes. But this argument is insufficient, even when the amount of contraband and the diversification of its supply is taken into account. The main reason is that laundering coca-dollars through this means is relatively less expensive. The conditions for the efficient functioning of this mechanism are simple: first of all, there must be a significant informal sector, notably in commercial activities; and secondly, there must also exist a free zone. This is the case of Colon in Panama. Criminal organizations buy merchandise in the free zone, paying for it in cash or in partially laundered money, sometimes using letters of credit (the controls being fewer, even non-existent in the free zones). This merchandise is then transferred into contraband in Colombia, where it is sold in small shops, which are called "San Andreas"-named after a Colombian island historically famous for this activity, Laundering passes, therefore, through an activity of contraband and then through an illegal commerce that is far from marginal: the "San Andreas" constitute a veritable network, consisting some-times of supermarkets where one finds extremely diverse products at competitive prices ( CID: 1997). The sums of laundered money are significant: approximately 1.3 billion dollars in 1993 and in 1994, which is significantly more than that of 1991 (327 million) and 1992 (634 million).^[18]
 

What remains to be discussed, is the international financial markets, De Maillard ( 1998) showed how the deregulation of these markets allowed for a rise in criminal finance. The techniques that are increasingly used are the following: wide-spread over- and under-billing, transferring of funds from one account to another, using offshore locations, certain banks practicing secret or double book-keeping, investing in short-term high-risk products, giving a legal status to sums received, and finally a plethora of avenues to repatriate capital^[19]
 

The use of all of these techniques is becoming increasingly expensive. One would have assumed that financial liberalization, the rise of offshore financial institutions, and the development of emerging stock ex changes would lower the cost of these transactions. In fact the opposite has taken place. The increasing complexity and sophistication of financial instruments certainly allows for the conveyance of capital in ways that are opaque and therefore allow it to be laundered, or rather to be given a legal status. But the ensemble of these operations is expensive. Observers agree that the costs of laundering surpassed 5% to 8% in the mid-1980s and reached 15% to 20% at the end of the 1990s (Steiner 1997: 38-39).

A MACRO-ECONOMIC ESTIMATE OF REPATRIATING

Determining a macro-economic estimate of repatriating is difficult for the reasons we have already noted, but also because the dirty money that is repatriated and laundered is not limited to the money from drugs alone, In general, the international organizations consider that half of all laundered money comes from the traffic of illegal drugs. In certain countries, this percentage is considerably less, as gambling, weapons, and in particular , prostitution are responsible for a large majority of laundering operations. According to the data gathered by G. Fabre (1988: 77 ff.), arms trafficking, procuring, hydrocarbide smuggling, clandestine gambling, and narcotics and labor trafficking bring in 24 to 32 billion dollars a year in Thailand, which is a sum equivalent to the budget of the state. The traffic of narcotics alone for Thailand is estimated at a billion dollars and would thus constitute a min or activity. Conversely, in Colombia, it seems obvious that laundered money comes essentially from drugs. This is the case, but it would nonetheless be an error to think that the traffic of narcotics constitutes the sole component of laundered funds. Colombia produces emeralds and sells a large portion of them clandestinely (Guillelmet 1998).^[20] Guillelmet estimates the illegal sale of germs in Colombia at approximately 10% of the amount of exports-this figure being probably lower in the years from 1993 to 1995-that is, some 700 to 800 million dollars.
 

For our purposes, by limiting the estimates for laundering to the traffic of narcotics only, the estimate of the sums laundered passes through a simple calculation whose terms are known within a more or less significant margin of error. Grass earnings are the result of the quantity of successful exports-in other words, total exports minus confiscations-times the estimated average gross price, that is, 17,600 dollars per kilo in 1990. It is necessary to subtract from this figure of gross earnings gains-here 17,600 dollars for one kilo-the total costs resulting from this activity. Steiner's approach (1997: 38 ff.) is interesting: it is based on a separation between costs and earnings. It consists of subtracting from the gross earnings, the costs of conversion, corruption, and transportation, and the net earnings obtained in this calculation are then taken as equivalent to payments to the Colombian farmers, workers, and exporters. This requires a brief description. The transport costs of base cocaine in Bolivia and Peru -areas of production- are 100 dollars per kilo and those corresponding to the transport of cocaine from Colombia to the United States are 3,000 dollars a kilo, 50% of which is paid directly in cash. The cost of transport to Europe is considered to be 30% higher. By counterbalancing the destinations by the size of the markets, one obtains 3,100 dollars per kilo as the average cost for the transport of cocaine. The conversion from base into cocaine is accomplished through the use of chemicals whose cost is estimated at 200 dollars per kilo of cocaine produced (some estimates make reference to greater sums). The dirty money must also be laundered. We have already noted that the cost of this operation has increased greatly from the 1980s to the present. It is estimated at between 15% and 20% of the sums to be laundered. Steiner sets the figure at 10% until 1989 and 20% of net earnings thereafter. Finally, an extra $500 per kilo of cocaine can be added to the total costs, representing the sums spent to corrupt, to buy silences, and so on, As we have already indicated, the average wholesale price of approximately one kilo of cocaine is 17,600 dollars. At retail, this price increases on average to 130,000 dollars per kilo, while one kilo of base (expressed in the equivalent of HCL) was 500 dollars in Peru and 700 dollars in Bolivia, or about 600 dollars on average. The total cost of transportation (from the heart of the Andes to the United States), conversion, corruption, and laundering increases to 6,800 dollars per kilo, which is a little less than 40% of gross earnings per kilo. The remaining 60% serves to finance the payment of farmers, chemists, and the ensemble of Colombian mafia implicated in the wholesale traffic of cocaine. The Mexican mafias, who transport a substantial part of cocaine (50% to 70% according to the official estimates of 1996), receive a significant part of the earnings that are posted as the costs of transport. The sums acquired during this operation are laundered by these criminal organizations and are therefore not accounted for in those that the Colombian mafias have to launder, Moreover, the growing participation of Mexicans in the traffic of cocaine and the payment of a significant portion directly in cash actually drastically reduces net Colombian earnings as we have calculated them, by increasing part of the transport costs and by correspondingly reducing part of the Colombian exporters' price. The estimated net earnings of the Colombians are probably therefore overestimated, all the more so to the degree that a growing part of base-cocaine is today converted in Bolivia and transported by new routes, especially Brazilian ones ( Geffray 1997, 1998). Be that as it may, the estimations of the net gains of laundering obtained from 1987 to 1995 are on average 1.638 billion dollars with a minimum of 1.2 billion in 1994 and a maximum of 2.5 billion in 1989.
 

When we add to these net gains those that are acquired from the production of exported marijuana and from the recent production of heroin ( with the assumption that it is the wholesale prices at embarkation of the latter that are taken into account), we obtain approximately 2.5 billion dollars to which is added the laundered sums drawn from the illicit traffic of emeralds, that is 600 to 700 million dollars net of laundering fees. The results of these estimates, with the exception of emeralds, are presented in Table 4.
 

The laundered sums are considerable. In relation to official exports, they reach significant proportions: 35% in 1992, 34% in 1993, 27% in 1994, and 24% in 1995 for the laundering alone of money from the traffic of narcotics. The trend is certainly on the decline, because of the opening up of the economy and the strong growth of legal exports starting in 1994, but it remains at a very high level. This being the case, it is evident that from a strictly macro-economic point of view, this inflow of dollars in the diverse forms adopted by laundering, is not without influence on economic activity in general, One could speculate for example, that just like royalty payments, it could cause a "Dutch disease," in other words it could determine the rate of ex change, and participate in the destruction of entire sides of the economy because of a lack of competition following a differentiation of relative prices between exposed and protected sectors. Such a development is not, however, necessarily inherent in the logic of the narcotics industry (Salama 1994). It is problematic to attribute to the cultivation, conversion, and exportation of illicit drugs, the term royalty insofar as, on the one hand, it is a question of reproducible activities in contrast to black gold ( oil), for example; and on the other hand, it deals with private illegal activities on which, by definition, the state cannot collect taxes. The only comparison that can be made with a royalty is that the earnings coming from this illicit activity do not depend on work, but on a prohibition. Like mining royalties, enrichment is not the product of an ability to exploit the work force in an efficient manner, but of the possibility of establishing a position in the flow of income. This being the case, the considerable sums acquired from this activity would provoke an appreciation of the national currency, This was noted in the 1980s in Colombia when the unofficial rate of ex change was appreciated in relation to the official rate of ex change, in contrast to what was observed at the same time in most Latin American countries, Conversely, the recent changes in ex change rates in the Andean countries is not oriented towards an appreciation, but rather on the contrary, numerous countries have had to de value their currency because of the Asian-Russian crisis of 1997-1998, Several factors can in fact thwart the possible effects of an inflow of narco-dollars: a trade deficit following tariff remission, a growing instability in the balance of current funds due to national debt and dividend payments, or a budget deficit.^[21] Hence generalizations about the effects of the infusion of narco dollars into an economy must be treated with reservations.

Table 4. Estimates of Net Earnings, Colombian Drug Trade (in Millions of Dollars)
 

Source: Steiner (1997.48);
*GSM for Gomez and Santa Maria (1994). "La Economia subterranéa en Colombia," in Steiner;
**Rocha in Thoumi 1997.
 

CAN THE BEHAVIOR OF THE MAFIAS CHANGE?

This final section does not presume to deal with the entire range of Mafia behavior, but rather to outline certain of its characteristics. The main question is to determine if Mafia organizations can behave as ordinary businesses, or on the contrary, if they will remain profoundly marked by their origins. This question cannot be answered in any simple way. Several factors must be taken into account: the importance of these organizations in the network, the organization of this network in the form of rings or cartels, the techniques of laundering-recycling, and finally the weight of the past, that is to say, the temporal factor not in its future dimension but in that of its past

The first two factors are important. In the literature, one often finds references to the relative weakness of criminal organizations and to their articulation in networks, The activity of production is limited since it is not very open to scale economies, both at the level of the cultivation of poppy or coca leaf and of their conversion. For this reason, the size of the businesses depends less on the search for such scale economies than it does on the objective of reducing maximum risks ( Cartier Bresson 1997). Therefore, the size of the economic enterprise will not be the same, depending on whether analysis is located at the level of production, conversion, sale at wholesale price, or finally sale at retail price. It can be surmised that if criminal organizations seek to integrate production, conversion, and whole-sale commerce, they will have neither the same dimensions nor the same , organization in the form of networks as those organizations that buy illicit substances at wholesale price and resell them afterwards through a chain of intermediaries that leads to the consumer. Neither the material problems encountered, nor the information regarding the risk, nor the possibilities to circumvent this risk by corruption are identical at these levels. It continues to be the case that these organizations are obviously unstable, because former con tracts may lead to duplicity-without a neutral authority able to. arbitrate-the merchandise is in substantial part confiscated, and the criminal hierarchy can be dismantled (Reuter in Cartier Bresson 1997: 79). But these risks are different according to the place occupied in the network 's hierarchy. Integration of vertical levels, excluding the organizations involvement in the stage of retail sale, would argue for a significant increase in size. But the risks incurred and the limited flexibility of a big organization lead to both a limiting of the size of given organizations and to the structuring of related organizations in the form of networks. In Latin American countries, it can be suggested that the criminal organizations operate in the shape of an eight: at the base is the organization and the surrounding farmers under contract, who produce the raw material; then come smaller units engaged in conversion, which is under the control of the criminal organization itself. This segment sells the drugs at wholesale, launders the profits, and then recycles them. A large base: the farmers; a large peak also: the retailers; and linking the two, a knot: the criminal organization. These different activities are both completely separated and joined together by the presence of the head of the mafia organization and his associates (Rocha in Thoumi 1997: 163).
 

Laundering-recycling is a very important activity of the criminal organization. It should not be confused with that of laundering-repatriating, even though the two often overlap, and even merge with one another. Recycling is in general facilitated by the existence of a weak state whose administrations and other apparatuses are strongly susceptible to corruption and to the existence of a sizable informal economy, which in contrast with the traffic of narcotics, illegally produces goods and services whose production and consumption are not prohibited, The existence of this informal economy in Colombia and the constitution of the state marked by its recent past ( a role played by violence, exclusion, and effectively weak citizenship) allow for the blurring of the lines between the illegal and legal (Rivelois 1999: 199) and ultimately authorize the rise of illicit activities.
 

The expenditures of drug dealers generate few jobs and little wealth. The importance of the effects of these expenditures on economic growth depends on a number of factors, If it is a question of extravagant or directly speculative expenditures, such as the purchase of land, the effects on the creation of wealth are few, if not non-existent When it is a question of spending and investment in the construction sector, the effects on upstreamtraining can be significant as well as a source of new productive activities due to the expansion of upstream markets, The indirect effects on employment and the creation of wealth depend, therefore, on both the portion devoted to investment in the expenditures of drug dealers and the places where the investments are carried out, The most speculative expenditures generate few jobs, except in construction, those that are less speculative participate in the creation of jobs according to the size of the investment, the techniques used and above all, the possible effects on upstream training.

Recycling takes place primarily in certain sectors ( Castelli 1999) like tourism (restaurants, hotels, casinos), because by its very nature it opens the way to future launderings; in real estate speculation and the purchase of land (where regulations concerning the source of funds are in general more lax and the possessor of laundered money can recycle and find a status for this money, allowing him to run fewer risks of investigations of the origin of these funds); in the pharmaceutical industry (allowing for the acquisition of the chemicals necessary for the conversion of narcotics raw materials, without too many risks); businesses located in sectors where there is considerable possibility either to falsify accounts-under- or over-billing-or to set up double accounts; or in the financial service sectors (banks, stock companies).
 

The panoply of businesses at the margin of directly illicit activities has two logics, one is the classic reproduction of capital, the other is the laundering or recycling of money. These two activities are complementary to a certain degree, and it would be erroneous to think that the first can entirely take the place of the latter, since they rest upon two opposing ways to resolve conflicts: law and violence. The two cannot, however, coexist at length without informing each other: either the business loses its mafia status, or it conserves it-and as a result the law grows gangrene in its application by the violence of corruption or direct physical violence.
 

The weight of the past is not neutral It is an error to think that behavior is ordered by the sole optimization of choice aimed at its future objectives (Dupuy 1997). Past error cannot be treated as an "irrecoverable cost" that should be accepted in order to optimize the choices that aim at a possible future. It affects behavior and therefore influences subsequent choices, as if it were a question of "redeeming" the error made.^[22] This observation on rationality seeks to explain why when an increase in wealth comes from an activity of illicit income as well as from an ability to fix a position in the flow of this income, it is very difficult to abandon this manna in favor of an increase of wealth that is less lucrative and which comes from the organization and exploitation of the work force, in other words, from profit ma king. In the same way, it is very difficult to aim for the long-term in the choice of investments.^[23] The "error" tends to repeat itself and the illegal tends to gain ground over the legal, transforming businesses into mafia businesses and making the "nobilization " of criminals within one generation very difficult.
 

We conclude by reemphasizing that, contrary to what one might think, a reliable quantitative estimation of cocaine traffic is possible, even though we are working within an economy of information that is at best imperfect, The simplest techniques consist of combining the estimates of production and consumption. This estimate is possible because the United States is the principal consumer and Colombia is the principal producer of cocaine, This would be much more difficult to accomplish if we were not in the presence of this particularity. The reliability of this estimation therefore rests in large part on the estimation of production alone. This estimate can be confirmed by the estimates of repatriation, The macro-economic estimates certainly rest upon assumptions that often seem overly bold-we do not account for the presence of Colombian networks in wholesale and retail trade; and the growing role of Mexican criminal organizations seems neglected, as does the impact on the volume of business of new routes in Argentina and Brazil But despite these difficulties and insufficiencies, this macro-economic estimate is relatively reliable in the sense that it talleys with the findings we have been able to obtain based on the estimates of different techniques used to repatriate and launder money from criminal activities. It is pertinent to emphasize the degree to which these estimates differ from those made "for political use" in order to justify this or that means of violent reprisal, not to speak of their diffusion to legitimate and defend a budget, or on behalf of an organisation in charge of fighting the war against the illicit traffic in drugs.

[1]. Translated by Nanette Lynne Fornabai.
[2]. To borrow the expression used by Steiner (1997).
[3]. Such as the commodification of sports and in a more general way. the stress linked to the need to reach certain levels of performance and, short of that. the often legitimate fear of losing.
[4]. One need only to think of the significant proportion of the population heavily dependent on the strongest and most diverse sedatives.
[5]. Fabre 1998.
[6]. Machado 1997; Steiner 1997; Thoumi 1997.
[7]. See Camara 1999.
[8]. Thoumi et al. 1997.
[9]. Franks 1991 in Steiner 1997: 18.
[10]. Geffray 1997.
[11]. 1t is necessary in fact to deduct from this production the local consumption of cocaine, which in the large cities, tends to be growing.
[12]. Rivelois 1997; Dupuis 1998.
[13]. This represents the minimum price paid in Miami. Until the end of the 1980s, we only considered this price. Since then, we have taken into account the weak but growing participation of Europe (approximately 10%~~ of the market), where the prices are almost double the prices offered in Miami.
[14]. According to the United Nations International Drug Control Program, the total sale of all drugs is estimated between 400 and 500 billion dollars. an estimate that M.C. Dupuis (1998) borrows likewise without discussion. only to give another. different figure a few pages later, closer to the estimate that we have borrowed. The retail sale of heroin would be on average 17 billion dollars and the sale of cocaine 30.5 billion in the United States, that is, less than 50 billion dollars (Dupuis 1998: 48). To this number, it is necessary to add certainly the consumption outside of the United States, and as such we are far from so-called "folkloric.. estimates.
[15]. When we compare the sums sent by Colombians residing in the United States to Colombia before 1980 and after. we observe more than a significant increase. The explanation for such increase must be the improvement of their standard of living (for further details. see Rocha in Thoumi 1997. 193 ff. ).
[16]. 1t is necessary to know that the sums attributable solely to the traffic of cocaine are estimated at approximately 11,200 tons of 5, 10, and 20 dollar-bills. Even when changed into 100 dollar-bil1s and limited only to the wholesale price, their influence remains considerable (see Dupuy 1997).
[17]. Conversely, if there are difficulties in exporting the capital, then the over-billing of imports becomes advantageous. One can also use under- or over-billing on the price of exports. But this method is difficult when the exports of the country are principally composed of raw materials whose price is fixed internationally.
[18]. The estimate, approximate but reliable, obeys a simple principle. it is sufficient to note the amount of re-exportation in the free zone of Colon in the direction of Colombia, provided by the zone services; subtract from these figures the amount of imports coming from the free zone and Panama provided by the Colombian statistical service (above all. the statistics from 1MF should not be used because they do not take into account the figures from the free zone, but only those from Panama-which maintains only a marginal commerce with Colombia). The information obtained reveals both the technique of under-billing and a significant amount of contraband in the open, which finds its natural outlet in the network of the "San Andreas" shops.
[19]. Obviously, it is difficult to account for the movement of capital caused by "normal" activity. The calculation of "excessive" movements is based on a reformulation of the hypothesis that these movements would not be greatly influenced by the changes in the rates of interest and exchange, but rather influenced by the production of illicit substances (for further information, see Steiner 1997; Urrutia and Ponton 1993). 1t is estimated at 600 million dollars on average for 1985 and 1989, at 1.17 billion from 1990 to 1992, and a little more than 800 million in 1993 and 1994 (Steiner 1997.68). It is evident I y difficult to attribute to narcotics traffic alone the responsibility of these "excessive" movements, since other illegal activities such as the traffic of emeralds exist in Colombia.
[20]. 70% of emeralds are exported illegally. The difference between the recorded and posted exports of the country and the recorded imports from outside the country is sometimes significant. Weak in Japan, the relationship between the entrances and the exits reaches more than 80% in the United States, 92% in Switzerland, these three countries totaling 80% of the foreign demand of Colombian gems (Guillelmet 1998.261 ff.). These percent-ages vary with time, depending on the changes in the selling price of emeralds, legislation on taxes; the connections with the traffic of narcotics, and so on. It is estimated. for example, that at certain times, the legally exported portion grows and is accompanied by an over-billing of exports to allow for the laundering of a part of drug money (Guillelmet 1998:251).
[21]. Let us recall an accounting proof the net surplus of the balance of current accounts should be equal to the net surplus of the private savings on the investment of residents to which should be added the net surplus of the budget. This net surplus is equivalent to the accumulation of the net assets in foreign countries.
[22]. Let us give a classic example. when the purchase of a second home is considered as having been an error. The best thing would be to sell it and spend one's vacation in a hotel. However, we note that people having committed this error , will spend their vacation in this home, as if they want to diminish the cost of that purchase, which apparently, according to the classic criteria of rationality is completely irrational.
[23]. Thus the preference for high-risk bonds when activities are oriented towards speculation at the Stock Exchange.

REFERENCES

Camara, M. 1999. "L'émergence des drogues en Afrique: quelques pistes méthodologiques pour identifier les liens licite/illicite," Revue Tiers Monde. 159, Paris. PUF.
Cartier Bresson, J. 1997. "Etat, marchés, réseaux et organisations criminelles entrepreneuriales," in Criminalité organisée et ordre dans la société. Aix en Provence. Presses Universitaires d'Aix-Marseille.
Castelli, B. 1999, "Impacts urbains du recyclage de l'argent de la drogue dans la région des Andes. une évaluation critique" Revue Tiers Monde: 159. Paris: PUF.
CID. 1997. "Dinamica comercial y lavado de dolares de los sandrenistos," in Informe final. Bogota: Universidad Nacional de Colombia.
De Maillard, J. 1996. "Le crime à venir, vers une société fractale." Le Débat 92.
De Maillard, J. 1998. Un monde sans loi, la criminalité financière en images. Paris. Editions Stock.
Dupuis, M.C. 1998. Finance criminelle, comment le crime organisé blanchit l'argent sale. Paris. Presses Universitaires de France.
Dupuy, J.P. 1997. "Temps et rationalité: les paradoxes du raisonnement rétrograde," in Les limites de la rationalité, rationalité, éthique et cognition. eds. J.P. Dupuy and P. Livet Paris. La Découverte.
Fabre, C. 1998. Les prospérités du crime. trafic de stupéfiants, blanchiment et déstabilisation financière dans l'après guerre froide. Paris. Publication in process.
Geffray, C 19911 and 1998. "Trafic international, blanchiment local et politique" and "Effets sociaux. économiques et politiques de la pénétration du narcotrafic en Amazonie brésilienne." ORSTOM. 3 and 4.
Guillelmet, J.M. 1998. "L'économie informelle comme mode de développement institutionnalisé. une étude à travers du cas pratique de la filière de l'émeraude colombienne." Ph.D Dissertation. Université de Nice-Sophia Antipolis
Kopp. P., ed. 1995. I. 'économie du blanchiment. Paris. Association d'Economie financière. Kopp, P. and M. Schirray, eds. 1994. Géopolitique et économie politique de la drogue. Paris. Furturibles. See in particular the articles of M. Schirray, "Les filières-stupéfiants. trois niveaux, cinq logiques" and "Les stratégies de survie et le monde des criminalités."
Machado, L.O. 1997. "Les mouvements d'argent et le trafic de drogue, une approche régionale." Photocopies.
Reuter, P. et al. 1993. Comparing Western European and North American Drug Policies. Drug Policy Research Center: RAND.
Rivelois, J. 1999. "Drogue, corruption et métamorphoses politiques, application à une comparaison Mexique-Brésil," Revue Tiers Monde. 159. Paris. PUF
Rivelois, J. 1997 Prince des paradis. pouvoir, drogue et corruption depuis le Mexique: Paris. Publication in process.
Salama, P. and M. Schirray, eds. 1992. Drogues et développement. Paris: Presses Universitaires de France. See in particular the studies of G. Fonséca, "Economie de la drogue: taille, caractéristiques et impact économique"; P. Kopp, "La structuration de l'offre de drogue en réseaux"; and B. Destremeau, "Les enjeux du quat au Yémen."
Salama, P. 1994. "Drogues et économie dans les pays andins, approches méthodologiques," Tiers Monde 137. Paris. Presses Universitaires de France.
Steiner, R. 1997. Los dol/ares del narcotrafico. Cuadernos Fedesarrollo 2.
Thoumi, F.E. et al. 1997. Drogas ilicitas en Colombia, su impacto económico, politico y social. Bogota. Ariel. See in particular F. Thoumi, "Introducción y panaroma"; S. Uribe Ramirez, "Los cultivos ilicitos en Colombia"; R. Rocha Garcia, " Aspectos económicos de las drogas ilegales"; and E.A. Garzon Saboya, " Aspectos legales y praxis del narcotrafico y lavado de dinero"
Thoumi, F.E. 1994. Economia politica y narcotra~fico. Bogota. Tercer Mundo.
Urrutia, M. and A. Ponton. 1993. "Entrada de capitales, diferenciales de interés y narcotrafico," in Macroeconomia de los flujos de capital en Colombia y América Latina, eds. M. Cardenas and L.J. Garay. Bogota. Tercer Mundo, Fedesarrollo Fescol.