The Economist

"IF ONLY it were legitimate, there would be much to admire about the drugs industry. It is, to start with, highly profitable. It produces goods for a small fraction of the price its customers are willing to pay. It has skilfully taken advantage of globalisation, deftly responding to changing markets and transport routes. It is global but dispersed, built upon a high level of trust, and markets its wares to the young with no spending on conventional advertising. It brings rewards to some of the world’s poorer countries, and employs many of the rich world’s minorities and unskilled.

However, it is an odd business. Its products, simple agricultural extracts and chemical compounds, sell for astonishing prices. A kilo of heroin, 40% pure, sells (in units of less than 100 milligrams) for up to $290,000 on the streets of the United States?enough to buy a Rolls-Royce car. These prices directly reflect the ferocious efforts by the rich countries to suppress drugs. The effect is to drive a massive wedge between import and retail prices. The import prices of both heroin and cocaine are about 10-15% of retail prices in rich countries. In poor countries, the ratio may be more like 25%. Add a little more for seizures, valued at import prices, and the grand total is probably about $20 billion. That would put the industry in the same league as Coca-Cola’s world revenues.

Taken at retail prices, it is almost certainly the world’s largest illicit market, although probably smaller than the widely quoted estimate by the United Nations Office of Drug Control and Crime Prevention of $400 billion, which would put it ahead of the global petroleum industry. Every number about the production, consumption and price of drugs involves much guesswork, a warning that applies all through this survey. But global retail sales are probably around $150 billion, about half the sales of the (legitimate) world pharmaceutical industry and in the same league as consumer spending on tobacco ($204 billion) and alcohol ($252 billion).

The estimate of world drug sales comes from Peter Reuter, an economist at the University of Maryland and co-author (with Robert MacCoun) of a comprehensive new study of illegal drugs on which this survey frequently draws. He notes that the official estimate of retail drug sales in the United States is $60 billion, making America easily the world’s most valuable market. European sales are at most the same again, probably less. Pakistan, Thailand, Iran and China account for most of the world’s heroin consumption, but prices are low, and so sales in total are probably worth no more than $10 billion. Add in Australia and Canada; add, too, Eastern Europe and Russia, where sales are growing fast, but probably still make up less than 10% of the world’s total. Exclude European marijuana, much of which is domestically produced.

It may seem distasteful to think of drugs as a business, responding to normal economic signals. To do so, however, is not to deny the fact that the drugs trade rewards some of the world’s nastiest people and most disagreeable countries. Nor is it to underestimate the harm that misuse of drugs can do to the health of individuals, or the moral fury that drug-taking can arouse. For many people, indeed, the debate is a moral one, akin to debates about allowing divorce, say, or abortion. But moral outrage has turned out to be a poor basis for policy.

Nowhere is that more evident than in the United States. Here is the world’s most expensive drugs policy, absorbing $35 billion-40 billion a year of taxpayers’ cash. It has eroded civil liberties, locked up unprecedented numbers of young blacks and Hispanics, and corroded foreign policy. It has proved a dismal rerun of America’s attempt, in 1920-33, to prohibit the sale of alcohol. That experiment?not copied in any other big country?inflated alcohol prices, promoted bootleg suppliers, encouraged the spread of guns and crime, increased hard-liquor drinking and corrupted a quarter of the federal enforcement agents, all within a decade. Half a century from now, America’s current drugs policy may seem just as perverse as Prohibition.

For the moment, though, even having an honest debate about the policy is extremely difficult there. Official publications are full of patently false claims. A recent report on the National Drug Control Strategy announced: ?National anti-drug policy is working.? In evidence, it cited a further rise in the budget for drugs control; a decline in cocaine production in Peru and Bolivia (no mention of Colombia); and the fact that the proportion of 12th-grade youngsters who have used marijuana in the past month appears to have levelled off at around 25%. If these demonstrate success, what can failure be like?

Nearer the truth is the picture portrayed in ?Traffic?, a recent film that vividly demonstrated the futility of fighting supply and ignoring demand. In its most telling scene, the film’s drugs tsar, played by Michael Douglas, asks his staff to think creatively about new ideas for tackling the problem. An embarrassed silence ensues.

This survey will concentrate largely (but not exclusively) on the American market, partly because it is the biggest. Americans probably consume more drugs per head, especially cocaine and amphetamines, than most other countries. In addition, the effects of America’s misdirected policies spill across the world. Other rich countries that try to change their policies meet fierce American resistance; poor countries that ship drugs come (as Latin American experience shows) under huge pressure to prevent the trade, whatever the cost to civil liberties or the environment.

Moreover, America’s experience demonstrates the awkward reality that there is little connection between the severity of a drugs policy on the one hand and prevalence of use on the other. Almost a third of Americans over 12 years old admit to having tried drugs at some point, almost one in ten (26.2m) in the past year. Drugs continue to pour into the country, prices have fallen and purity has risen. Cocaine costs half of what it did in the early 1980s and heroin sells for three-fifths of its price a decade ago. Greater purity means that heroin does not have to be injected to produce a high, but can be smoked or sniffed.

However, American experience also suggests that the pattern of drug consumption is altering, arguably for the better. Casual use seems to have fallen; heavy use has stabilised. More American teenagers are using cannabis (which, strictly speaking, includes not just the herb?marijuana?but the resin), but the number of youngsters experimenting with cocaine or heroin has stayed fairly steady. The American heroin epidemic peaked around 1973, since when the number of new addicts has dropped back to the levels of the mid-1960s. The average age of heroin addicts is rising in many countries?indeed, the Dutch have just opened the first home for elderly junkies in Rotterdam. America’s hideous crack epidemic has also long passed, and cocaine use has retreated from its 1970s peak. And a recent study shows that the likelihood of proceeding from cannabis to harder drugs such as cocaine or heroin has fallen consistently for a decade. ?We are largely dealing with history,? says Mr Reuter. ?The total population of drug users has been pretty stable since the late 1980s.?

This is not an unmixed blessing: heavy users seem to be using more drugs, and to be injuring and killing themselves more often. As with cigarette-smoking, drug-taking is increasingly concentrated among the poor. And in some rich countries other than America, such as Britain, the number of both casual and heavy users of most drugs is still rising. In the poorer countries and in Central and Eastern Europe too, drugs markets are flourishing. India and China are probably the fastest-growing large markets for heroin.

But in the rich countries, the drugs that increasingly attract young users are those that are typically taken sporadically, not continuously: cannabis, ecstasy, amphetamines and cocaine. In that sense, they are more like alcohol than tobacco: users may binge one or two nights a week or indulge every so often with friends, but most do not crave a dose every day, year in, year out, as smokers generally do. That does not mean that these drugs are harmless, but it should raise questions about whether current policies are still appropriate.

Today’s policies took shape mainly in the mid-1980s, when an epidemic of crack cocaine use proved a perfect issue around which President Ronald Reagan could rally ?middle America?. His vice-president, George Bush, called for a ?real war on drugs?, which caught the mood of the time: opinion polls showed that drugs were at the top of people’s lists of worries. By the early 1990s the crack scare had faded, but a series of increasingly ferocious laws, passed in the second half of the 1980s, set the framework within which Mr Bush’s war on drugs is still waged today.

This framework is not immutable, although formidable vested interests?including the police and prison officers?now back tough drugs laws. Attitudes to policy change over time (see article), and drugs policies in many countries are changing with them. Governments are gradually putting more emphasis on treatment rather than punishment. Last autumn, in a referendum, California voted to send first- and second-time drug offenders for treatment rather than to prison. And the law on possessing cannabis is being relaxed, even in parts of the United States, where several states now permit the possession of small amounts of it for medical use.

In Europe and Australia, governments have relaxed the enforcement of laws on possessing ?soft? drugs. In Switzerland, farmers who grow cannabis for commercial sale within the country will be protected from prosecution if a new government proposal goes through. In Britain, Michael Portillo, a top opposition politician, advocates legalisation. But it is hard for an individual country to set its own course without becoming a net exporter, as the experience of Europe’s more liberal countries shows. Ultimately, the policies of the world’s biggest drugs importer will limit the freedom of others to act.

At the heart of the debate on drugs lies a moral question: what duty does the state have to protect individual citizens from harming themselves? The Economist has always taken a libertarian approach. It stands with John Stuart Mill, whose famous essay ?On Liberty? argued that:

The only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him, or visiting him with any evil in case he do otherwise. Over himself, over his own body and mind, the individual is sovereign.

This survey broadly endorses that view. But it tempers liberalism with pragmatism. Mill was not running for election. Attitudes towards drug-taking may be changing, but it will be a long time before most voters are comfortable with a policy that involves only remonstration and reason. People fret about protecting youngsters, a group that Mill himself accepted might need special protection. They fret, too, that drug-takers may not be truly ?sovereign? if they become addicted. And some aspects of drug-taking do indeed harm others. So a first priority is to look for measures that reduce the harm drugs do, both to users and to society at large.

VOTERS?and governments?change their minds about ways to deal with activities they disapprove of. Governments used to ban gambling; now many run their own lotteries. Prostitution, although still generally illegal, is rarely the target of police campaigns. Attitudes to alcohol have changed in the past century-and-half. So have attitudes to drugs.

In 19th-century America, campaigners talked of the demon drink in much the same way that they now talk of drugs. The temperance movement blamed booze for crime, ?moral degeneracy?, broken families and business failure. In America, this led to Prohibition, with its accompanying crime and bootlegging. In England, campaigners won restrictions on access, in the shape of the pub-closing hours that have puzzled foreign visitors ever since. It may have been a bore, but it was a less socially costly way of dealing with an undesirable habit than a ban.

Today’s illegal drugs were patent medicines in the 19th century. Morphine and opium were freely available in both Europe and America. Victorian babies were quietened with Godfrey’s Cordial, which contained opium. Cocaine was the basis of remedies for the common cold. When Atlanta prohibited alcohol, John Pemberton, producer of a health drink called French Wine Coca, developed a version that was non-alcoholic but still contained traces of coca, thereby creating the world’s best-selling soft drink. As for marijuana, Queen Victoria reputedly used it to soothe the royal period pains.

Far from opposing the drugs trade, the British and the Americans notoriously promoted it in the 19th century. In 1800 China’s imperial government forbade the import of opium, which had long been used to stop diarrhoea, but had latterly graduated to recreational use. British merchants smuggled opium into China to balance their purchases of tea for export to Britain. When the Chinese authorities confiscated a vast amount of the stuff, the British sent in gunboats, backed by France, Russia and America, and bullied China into legalising opium imports.

Initial efforts to stamp out drug use at home had little to do with concerns about health. One of America’s first federal laws against opium-smoking, in 1887, was a response to agitation against Chinese ?coolies?, brought into California to build railways and dig mines. It banned opium imports by Chinese people, but allowed them by American citizens (the tax on opium imports was a useful source of federal revenue). The drafters of the Harrison Act of 1914, the first federal ban on non-medical narcotics, played on fears of ?drug-crazed, sex-mad negroes?. And the 1930s campaign against marijuana was coloured by the fact that Harry Anslinger, the first drug tsar, was appointed by Andrew Mellon, his wife’s uncle. Mellon, the Treasury Secretary, was banker to DuPont, and sales of hemp threatened that firm’s efforts to build a market for synthetic fibres. Spreading scare stories about cannabis was a way to give hemp a bad name. Moral outrage is always more effective if backed by a few vested interests.

THE drugs industry is simple and profitable. Its simplicity makes it relatively easy to organise; its profitability makes it hard to stop. At every level, its pricing and its structure are shaped by the high level of risk from enforcement: the risk of seizure and jail, and the uncertainty that arises because traders cannot rely on the law to enforce their bargains.

The industry’s products are of two sorts. Most of its products are agricultural, but a growing sideline is made from simple chemicals. Production of farmed drugs is concentrated increasingly in two countries: about two-thirds of the world’s heroin (made from opium) may come from Afghanistan and most of the rest from Myanmar; four-fifths of coca from Colombia. Only cannabis is produced in large quantities not just in the poor world?principally Mexico?but also in the rich, where much of the best stuff is grown. It is a tolerant crop. It can be interplanted in cornfields in Kentucky, or lovingly tended in an apartment in Amsterdam, where a taxi driver told this correspondent that he regularly raised 150 plants in a cupboard to sell at nine weeks for 60 guilders ($23) apiece to a local coffee shop. The bulkiness of cannabis, and its relatively low value, make it a crop best grown near the market.

Tracking crops is difficult, but easier, thanks to spy satellites, than tracking chemicals. Nobody is sure whether the Netherlands is the world’s main producer of ecstasy or (as seems more likely) merely the world’s main entrepôt for a product made in Poland and other parts of Eastern Europe. Methamphetamines seem to be produced mainly in small factories on both sides of the Mexican-American border. William Gore, of the Federal Bureau of Investigation in San Diego, thinks that successful law enforcement on the American side of the border has reduced factories there to making only 1-2lb (up to a kilogram) of the drug at a time; laxer vigilance to the south means that Mexican factories produce 100-200lb at a time.

Getting drugs from the poor world to the rich requires a distribution network. The task is tougher for cocaine than for heroin, because cocaine is more frequently shipped or flown to its markets, making it vulnerable to seizure. Most heroin consignments appear to travel overland. But this is where the big money starts to be made. The price paid to a Pakistani farmer for opium, reckons the United Nations, is $90 a kilo (see table, next page). The wholesale price in Pakistan is almost $3,000. The American wholesale price is $80,000. On the street, at 40% purity, the retail price is $290,000. As for cocaine, the leaf needed to produce a kilo costs about $400-600, according to Francisco Thoumi, author of a remarkable unpublished study of the Andean drugs industry. By the time it leaves Colombia, the price has gone up to $1,500-1,800. On America’s streets, after changing hands four or five times, the retail price for a kilo of cocaine works out at $110,000, and in Europe substantially more.

That vast gap between the cost of producing the stuff and the price paid by the final consumer goes a long way to explaining why drugs policies so often fail. However, the people who grow or make illegal drugs see only modest returns. The value is embedded mainly in the distribution chain. In Pakistan, for example, 90% of the domestic retail price of heroin goes to local wholesalers and retailers. The price at which heroin leaves the country may be only 10% of its street price in the United States or Europe.

Developing-country producers can find distribution difficult. Bruce Porter, the author of ?Blow?, a book about the 20-year career of a drug merchant called George Jung that has now been made into a film, recounts that the Colombians in the early 1970s had trouble getting their cocaine to the American market. ?George showed them how to distribute, using the marijuana distribution chain.? Once that was in place, ?George became a bulk transporter, shipping cocaine from Colombia to Colombians in Miami.?

The people who dominated the cocaine trade in Colombia in its early days were experienced smugglers, thanks to the country’s long history of gold and emerald smuggling. Much the same was true in Mexico, says Peter Smith, director of Latin American studies at the University of California, San Diego. When tough policing in Miami drove up their costs, Colombians formed joint ventures with Mexicans who were in the general smuggling trade, rather than with the small ?mom-and-pop? cartels in Mexico that had previously grown marijuana for sale in the north. They reckoned that the professional smugglers were more likely to have the logistics skills needed for the job.

In the early 1990s, these smugglers began to insist on being paid in drugs rather than cash, allowing them to break into American distribution too. They swiftly evolved from subcontracted transporters to urban distributors. The relationship is finely balanced: the Mexican smugglers know that, if they ask for too large a share, the Colombians can always return to shipping their cocaine by a different route.

Over the years, these distribution networks have become more efficient. That may explain one of the many mysteries of the drugs business: the halving of the price of heroin and cocaine between 1980 and 1990. The National Research Council speculates: ?The drug industry may have experienced the learning-curve effects often associated with new industries as they find ways to be more efficient in their operations.? In a footnote, the report adds: ?Learning by doing has a long history in studies of industrial organisation, productivity and growth.?

Certainly the Mexicans, according to a study done for the United Nations, seem to have concentrated on the drugs business in a way that might be expected to improve efficiency. Unlike other distributors, they avoid diversifying into other sorts of crime. Joseph Fuentes, a senior New Jersey policeman who has written a doctoral thesis on the industry, explains that the Mexican distributors operate with great professionalism, sometimes employing top managers with degrees in business studies, and relying heavily on honour, credit and collateral. ?The recruitment process is very like that for IBM or Xerox,? he says?except that the drug distributors require detailed information about the whereabouts of a prospective employee’s parents, spouse and children.

In Europe, distribution patterns seem to be different. The United Nations reckons that organised crime is less involved, at least in cocaine trafficking, and that more trade passes through ordinary businesses, many of them based in Spain. The retail side is often run by small groups or individuals supplying a network of friends; gang-controlled distribution is rarer. That may change: for instance, Martin Witteveen of the Dutch public prosecutor’s office believes that Israeli crime syndicates are taking over much of the trade in ecstasy between the Netherlands and the largest market, America.

Distribution within the rich importing countries is often dominated by immigrant groups. A police officer in Bern, in Switzerland, counts them off on his fingers: cocaine comes into the country mainly from Spain, but the trade is run by African asylum-seekers and by Turks. Heroin comes from Turkey and the Balkans, and the business is mainly in the hands of Albanians, Serbs and Macedonians, he says. Few of these folk appear in the streets: the final deal is often done by Swiss junkies. There are similar stories everywhere: in Denmark, it is Gambians, in Australia, Vietnamese.

This foreign control is no accident. Immigrant groups may have strong links with producing countries; they speak languages the police rarely understand; they have close ties of loyalty to each other. All these things give them a competitive advantage over locals. In addition, they have less to lose because they find it harder than locals to get decent legitimate jobs.

Given that heroin and cocaine are both highly concentrated, these dealing networks are probably not large: about 500 tonnes of cocaine come into the United States each year, and some dealers handle more than 10 tonnes a year. A few hundred people probably handle most of it.

The big battalions are on the streets. In poor parts of town, dealing is often a big source of employment. A study of drug markets in Milwaukee a couple of years ago by John Hagedorn, of the University of Illinois-Chicago, found that at least 10% of Latino and black men aged 18-29 drew at least part of their income from the drugs business. It was, he said, the most profitable activity in the town’s informal economy: 28 businesses, dealing mainly in cocaine, employed about 190 people, their owners grossing between $1,000 and $5,000 a month. Many of the owners also had jobs in the legitimate economy?drug selling seemed to be a complement to, rather than a substitute for, legitimate work. Thirteen of these businesses had been going for at least two years, developing innovative ways of avoiding the police and so reducing their business risk. The owners had stopped dealing from street corners or homes, and used pagers and mobile phones instead. They also employed runners to deliver drugs, and so carried almost no drugs themselves.

Different customers are willing to incur different risks. Richard Curtis of the John Jay College of Criminal Justice in New York, who has studied the retail market for drugs there, has found that customers in the smart areas of midtown and lower Manhattan tend not to travel to the shadier areas of Harlem or Washington Heights to buy drugs, even though they would save money if they did.

Recruiting employees appears to be easy. ?In a lot of poor communities, drug dealers are the only equal-opportunity employer,? says Deborah Small, director of public policy at the Lindesmith Centre, a drug-campaigning organisation. The main alternative source of illegal income, numbers betting, has been largely destroyed by legalisation. And drug-dealing pays well: one study of dealers in Washington, DC, at the height of the 1980s crack epidemic found that they could earn $30 an hour, compared with about $7 from legal employment.

That is an attractive rate, especially for the middle-aged high-school drop-out who is getting too old for mugging and has few other ways to make a living. But, as in every business, earnings vary with responsibility, and have to be set against the risks. A sophisticated study of the finances of one drug gang by two economists at the University of Chicago, Steven Levitt and Sudhir Alladi Venkatesh, found that, whereas the top members earned far more than their legitimate market alternative, the street-level sellers earned roughly the minimum wage. They seemed to stay in the job in the hope of rising to the top. But the risks are enormous: gang wars, essential to gain market share and to resolve disputes, also drive customers away?and for this particular sample resulted in a death rate of 7% among distributors.

Many of the ?runners? at the tip of the distribution chain are paid in a mix of drugs and cash. That turns drug-dealing into a sort of pyramid-selling, giving them an incentive to make more sales. And customers, as with any business, are the lifeblood of the drugs trade.

MOST drug users live in the poor world, not the rich. Countries such as China and Pakistan in the case of heroin, and Colombia (South America’s second most populous country) in the case of cocaine, have local traditions of drug use and vast uprooted urban populations to provide expanding markets. In future, growth will be concentrated in developing countries and the former Soviet Union.

At present, the markets with the big money are in the rich world, where the mark-ups between import and sales prices are highest. Here, not surprisingly, most people buy the drugs that have the fewest side-effects and are least likely to cause addiction. In that respect, drug users seem to behave as rationally as other consumers.

Everywhere, the most widely used drug by far is cannabis. At some point or another, about half the people under 40 in America have probably tried it. In time, as many adults in the rich world may have sampled cannabis as have tried alcohol. In many social groupings, especially in large cities, using cannabis has already become more or less normal behaviour. ?The last time anyone offered it to me,? recalls Paul Hayes, a senior British probation officer who has just become head of a new drug-treatment agency, ?was after a primary school parent-teacher association disco, in the home of a Rotary Club member, and the person was a detective-sergeant in the Metropolitan Police. If that’s not normalisation, I don’t know what is.? Prudently, Mr Hayes refused.

Other drugs are becoming part of the normal weekly pattern of life in some social circles. Amphetamines and cocaine, like cannabis, are mostly taken sporadically, and are used far more heavily by the young than by the middle-aged. Simon Jenkins, a former editor of the Times and member of an inquiry into drugs and the law under Lady Runciman, argues that London’s vibrant clubbing scene is clear testament to the profusion of drugs available there: how else would people have the energy to dance all night?

Most drug users, like those clubbers, are occasional dabblers. A 1997 survey of western German drug users sets the tone: just under 80% of cannabis users take the drug no more than once a week, and almost half take it fewer than ten times a year (see chart). With ecstasy and cocaine, users indulge even less often.

With drugs, as with alcohol, a minority of users tends to account for the bulk of consumption. In America, for instance, 22% of users account for 70% of use. Heroin use is probably even more dominated by frequent or dependent users. Most drug users, it seems, understand the risks they are taking, and approach them rationally. Of Europe’s adults, at most 3% are likely to have tried cocaine; fewer than 1% have ever sampled heroin.

Most drugs do not appear to be physically addictive. Views on this may eventually change: in laboratories all over the United States, unfortunate rats are being put into drug-induced hazes as the National Institute on Drug Abuse (NIDA) spends its hefty budget on a mass of research on the impact of drugs on the brain. Recent work on people who give up a heavy marijuana habit seems to show that they suffer anxiety and loss of appetite.

However, for the moment, the evidence suggests that neither marijuana nor amphetamines are physiologically addictive. Many people find it hard to abandon crack cocaine once they have tried it a few times, but when they do, they do not appear to become physically ill, as they would with heroin?or indeed nicotine or caffeine. ?Heroin is a true addiction, with a recovery rate of 40-50%,? explains Giel van Brussel, who has been head of Amsterdam’s addiction care department for many years. ?With cocaine, the recovery rate is around 90%, so we don’t see it as such an enormous problem.? That is rare sanity from a policymaker, but then Dutch policymakers are saner than most.

Even with the most addictive illegal drugs, only a minority of users seems to get hooked. With heroin, according to figures from America’s National Household Survey on Drug Abuse, one user in three is dependent. Alarming?but not compared with nicotine, which appears to be the most addictive drug of all: one study quoted by America’s Food and Drug Administration found that 80% of cigarette smokers were addicted (see chart 2, previous page). David Lewis, professor of alcohol and addiction studies at Brown University in Rhode Island, reckons that the relapse rates for those who try to give up are higher than those for heroin or crack cocaine. If the aim of drugs policy were to prevent harmful addiction, the main target of drugs enforcement agents would clearly be tobacco smokers and their dealers.

Studies of the routes by which people come to take up drugs have had a huge impact on policy. Most influential has been the ?gateway? theory, suggesting that soft drugs lead on to hard drugs: if cannabis is the path to crack cocaine, then clearly the sooner that path is blocked, the better.

In fact, this turns out to be nonsense. Certainly, most people who take ?hard? drugs have usually first smoked marijuana. But, as Lady Runciman’s excellent report on the misuse of drugs in Britain argued last year, for the ?gateway? theory to be proved correct requires not just that cocaine and heroin users are highly likely to have taken cannabis; it also requires that cannabis users are highly likely to move on to cocaine or heroin. Yet the vast majority of cannabis users do not graduate to these more dangerous drugs.

Moreover, there is no reliable evidence indicating that taking marijuana pharmacologically disposes people to later use of heroin. But work at Johns Hopkins University shows that children who drink and smoke in their early teens are disproportionately likely to progress later to marijuana. And a study in Britain found that the probability of 11-to-15-year-olds using an illicit drug is strongly related to under-age smoking and drinking. Beer and cigarettes seem to be gateways to marijuana, but marijuana does not seem to be a gateway to other drugs.

Whether somebody becomes a heavy drug user seems to depend on other factors. Heredity may play some part, and so may social conditions: recent American research has found that drug use is 50% more common in households that are welfare recipients than in those that are not. And family circumstances may interact with personality. Mr Hayes, after a long career in the London probation service, sees a typical user as ?someone who is a risk-taker?whose lifestyle involves bending rules.? Part of the lure of drug-taking seems to be the sense of danger. The question is how far people should decide for themselves whether to take such risks, and how far the government should make that decision for them.

IN A former warehouse under Manhattan bridge in New York, now home to a therapeutic community run by Phoenix House, 42-year-old Michael talks sadly of the cost of 30 years on heroin and crack cocaine. ?I couldn’t see further than the next bag of dope. I was hustling, shoplifting and getting high. I couldn’t deal with people. I have a 14-year-old son that I could never look after.?

Michael has now been voluntarily at Phoenix House for four months, learning how to cope with others and with himself. He is well-dressed, articulate and eager to escape. The desire to be a decent father to his son, coupled with the skills training and accommodation that Phoenix House provides, may be just what he needs to kick his habit, get a job and rejoin the human race. Gabriel has a tougher job ahead: he has been in a clinic in Tijuana, just south of Mexico’s border with the United States, after spending 14 of his 30 years on methamphetamines. His mother, who once threw him out for stealing from her, found him ragged and emaciated, and persuaded him to go for treatment at one of Mexico’s very few professionally run clinics. Now sleek and handsome, he is off drugs. But he will struggle to find work, and will return to live in the same community where his addiction began.

Many people take drugs because they get pleasure from them. To those who prefer a glass of burgundy and a cigar, that may seem hard to understand. It is, however, improbable that so many people would spend so much money on voluntarily eating, smoking or sniffing drugs if doing so brought them nothing but misery.

That said, though, abusing drugs undoubtedly wrecks many lives. Once people become truly dependent, it can take them years to break the cycle. As with cigarettes, the pleasure then consists mainly of avoiding the pain of giving up. But the vast majority of drug users end up like neither Michael nor Gabriel. They go through a period when drugs form part of their lives, and then they move on. Peter Cohen, of the Centre for Drug Research at the University of Amsterdam, followed a sample of cocaine users whom he describes as typical. After ten years, 60% had become completely abstinent and 40% remained occasional users. ?Most drug users ultimately stop,? he says. ?Drugs no longer fit their lifestyle. They get jobs, they have to get up early, they stop going to the disco, they have kids.?

The dangers of drugs should not be underestimated, but nor should they be exaggerated. With the exception of heroin, drugs contribute to far fewer deaths among their users than either nicotine or alcohol. In America, for instance, tobacco kills proportionately more smokers than heroin kills its users, and alcohol kills more drinkers than cocaine kills its devotees.

Consuming a drug is rarely the only cause of death. More often, the user is taking some extra risk. That is true even for heroin. The European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), which collects and analyses European statistics, reports that the mortality rate for people who inject heroin is two to four times as high as that for non-injectors, mainly because of the danger of contracting HIV or hepatitis from dirty needles.

?Acute deaths related solely to cocaine, amphetamines or ecstasy are unusual,? says the EMCDDA in a recent report, ?despite the publicity they receive.? Dr van Brussel, the addiction care expert, agrees: ?We have about 100 deaths a year from heroin addiction in the Netherlands,? he says, ?but only one or two from cocaine.? Even though much of the world’s ecstasy passes through the Netherlands, the country has only one death a year of a person with ecstasy in his bloodstream. Even then, it is rarely clear that ecstasy (MDMA) alone is the killer. According to Charles Grob, a professor at the UCLA School of Medicine, people who take MDMA incur health risks mainly if they are already unfit. He recalls one man, who had experienced no previous problems with ecstasy, whose blood pressure began to rise alarmingly after taking it. It transpired that he had stayed at a friend’s house, and used an asthma inhaler because he was allergic to his friend’s cat. It was the combination of substances that had caused the trouble.

Even drugs that do not kill people may still hurt them. More and more evidence suggests that drugs may affect brain activity. Some even hint that marijuana, regarded by its fans as safer than sugar doughnuts (and less addictive), may do damage. A study recently reported in the American Heart Association’s journal suggested that for middle-aged people the risk of a heart attack rose by nearly five times in the first hour after smoking marijuana.

But the overall impression remains that, in the words of the Lancet, a British medical journal, ?It would be reasonable to judge cannabis less of a threat than alcohol or tobacco…On the medical evidence available, moderate indulgence in cannabis has little ill-effect on health.?

Whereas some drugs harm people’s health, some may also do good. Hospitals still use heroin derivatives to treat pain. They cannot usually prescribe marijuana, even though a study published by the Institute of Medicine in 1999 suggested that marijuana could help to treat nausea, loss of appetite, pain and anxiety. In America, such findings have turned medical marijuana into the main issue in the campaign to soften the law on drugs.

Health apart, drugs cause other kinds of harm?not just to the individual user but to society at large. Crack cocaine seems to be linked to domestic violence, marijuana makes workers groggy, no drug is good for motorists. And some people who use drugs heavily??chaotic? drug users ? are disproportionately likely to commit crimes. A mere 5,000 of the country’s estimated 25,000 hard-drug addicts are responsible for about half of all petty crimes committed in the Netherlands, guesses Bob Keizer, drugs policy adviser to the Dutch Ministry of Health.

Given the expense of a heavy habit, petty crime is an obvious income source. However, Michael Hough, director of the Criminal Policy Research Unit at the University of the South Bank in London, believes that the link is not simple. Rather, the sort of person who becomes a ?chaotic? drug user is also disproportionately likely already to be an ?acquisitive offender?: a thief, shoplifter (the addict’s crime of choice) or burglar. ?The preconditions for starting on heroin are to be a risk-taker, and to have quite a bit of money,? he says. He points to a study of people arrested in Britain, by Trevor Bennett of Cambridge University, which calculated that the cost of consuming heroin and crack accounted for 32% of criminal activity.

Where drug use directly harms society, government is right to intervene. But the best way to protect society is not necessarily to ban drugs. If that were the right course, governments would begin by banning alcohol, which causes far more aggression and misbehaviour than any other substance, licit or illicit. Instead, governments everywhere pursue tougher policies against drugs, some of which are more harmful than the drugs themselves.

WHEN, 80 years ago, America prohibited the sale of alcohol, it imposed a milder policy than it currently applies to drugs, since people were allowed to possess alcohol for home use. Yet the 13-year experiment showed how easily a ban could distort and corrupt law enforcement, encourage the emergence of gangs and the spread of crime, erode civil liberties, and endanger public health by making it impossible to regulate the quality of a widely consumed product. The drugs war has achieved all these things but, since the business is global, it has done so on an international scale. In the United States particularly, and in those developing countries that supply it, the attempt to stamp out drugs has had effects more devastating than those of the drugs themselves.

The main targets of American supply-reduction campaigns over the years have been Bolivia, Peru, Colombia and Mexico. The net effect appears to have been a relocation and reorganisation of production, not a cutback. Dramatic falls in coca cultivation in Peru and Bolivia in the late 1990s coincided with an equally dramatic rise in Colombia, even though almost all the top people in Colombia’s notorious Cali cartel had been jailed in the mid-1990s. Estimates are sketchy, but the area under cultivation may have doubled. The decline in the price of cocaine in America has led the industry to look for new markets in Europe, and to diversify into the even more profitable opium.

Given the right conditions, it is clearly possible to suppress drug-growing in some regions. A country can shift the problem elsewhere, at least temporarily. However, the real factors that lead countries into or out of drug production seem to have much less to do with policy or prosperity than with culture and social institutions. As Mr Thoumi, author of the work on drugs in the Andes, points out, every country in the world that can produce bananas does so. Yet, in spite of a much larger gap between the export and import price of cocaine or heroin than of bananas, by no means every potential grower is in the business. He sees the explanation for Colombia’s booming business in its tradition of individualism, with few social controls. By contrast, Ecuador, a much poorer country that does not produce cocaine, has a stronger religious tradition.

If Mr Thoumi is right, government policy may have little durable impact on drug production. Basic economics suggest the same thing. Last year Congress voted $1.3 billion of emergency funding to Colombia to step up crop eradication over the next three years. But there are good reasons, spelled out in a recent article by Mr Reuter in the Milken Institute Review, why cutting off supply is doomed. The stuff is simply too profitable. Production is cheap. If a kilogram of cocaine retails for upwards of $110,000, the exporter can easily afford to double the few hundred dollars paid to the grower without much damage to his overall margin. Attempts to persuade growers to switch to planting pineapples are equally doomed: the cocaine exporters can readily outbid any reasonable scheme.

The same logic applies to shipping. American policy at the Mexican border concentrates on trying to stop the torrent of drugs that passes mainly through the Tijuana crossing, the world’s busiest border. But in Tijuana, once a dirt-poor town, drugs pay for smart new homes and cars. Some youngsters go to school with packets in their backpacks to sell at lunchtime.

The costs of seizure are small compared with the profits. Earlier this year, the US Coastguard seized two vast shipments of cocaine, one of 8 tonnes and the other of 13 tonnes. Together, they could have supplied 21m retail sales. To the astonishment of law-enforcement officers, the retail price of cocaine did not appear to budge. The enormous street value of the product makes it extremely cheap to ship. As Mr Reuter puts it, ?A pilot who demands $500,000 for flying a plane with 250 kilograms is generating costs of only $2,000 per kilogram?less than 2% of the retail price. Even if a $500,000 plane has to be abandoned after one flight, it adds only another $2,000 to the kilogram price.?

A profit margin such as this leaves enormous scope for corruption. Victor Clark Alfaro, a doughty human-rights campaigner in Tijuana, insists that: ?Corruption goes from the police on the street to the top officials.? The federal police, understaffed and underpaid on $700-800 a month, are no match for the big cartels. Francisco Ortiz Franco, an editor on Zeta, a newspaper that has had several run-ins with Mexican drug gangs, guesses that at least 20% of the agents fighting the drug trade are paid by the gangs; one dealer captured a couple of years ago put the figure for state and federal police officers at 80%. The problem is not that the police are particularly greedy: their option is usually to accept drug pay or risk retribution from the gangs.

Faced with such economics, the Bush and Fox administrations have been building closer links. For the first time, a big Mexican drugs boss was recently extradited to America to stand trial. And the American administration is at last willing to admit that?as President Bush said on a visit to Mexico earlier this year?the real problem is demand. But tackling demand is just as tricky as cutting off supply.

Superintendent Dean Ingledew of London’s Metropolitan Police is in charge of policing Soho, the city’s main nightclub district. His territory is full of Victorian alleyways, hostess bars and illegal drinking clubs. The customers who support Soho’s thriving crack trade are mainly ?rough sleepers?, homeless folk who can make up to ?100 ($140) a day begging in the street. But the market is changing: many more young professionals are coming in to sample a drug that has never before been popular in Britain, but now seems to be becoming more affordable.

Mr Ingledew and his colleagues use a mixture of community co-operation and street design, trying to improve lighting in Soho’s darkest nooks. They are developing ingenious ways to trap those dealers who keep their stock of ?rocks? in their mouths and swallow them when arrested. But ultimately their main goal is protecting public safety and the quality of life in Soho. Drug-dealing causes less disruption than belligerent drunks, but he is frank about the difficulty of tackling it. ?Our aim is to arrest the dealers,? he says, ?but there are a huge number of people who want to buy from them. So whenever we take a dealer out, the gap is filled. Enforcement is at best able to displace the market a few hundred yards, and to keep a lid on it.?

In New York, where the drug problem once bred horrific gang violence, the emphasis has been different. Michael Tiffany, deputy chief of the Bronx Narcotics Division at the New York Police Department, explains how putting a lot of officers into drug enforcement over the past eight years has brought successes. Up until 1994-95, he says, New York was the main distribution point for cocaine in the north-eastern United States. A decade ago, 50% of the people arrested for drug offences in the Bronx might have been from out of town. Now 95% of them are local. The wholesale distribution network has moved on.

Gone, too, has much of the violence. Bridget Brennan, special narcotics prosecutor for the City of New York, argues that increased enforcement has ?taken out the most disorganised?and most violent?organisations, that were shooting each other over spots. The ones left are more careful. They have a business interest in keeping violence down and not attracting attention to themselves.?

Her fear is that, with the violence gone, public support for tough policing may fade: ?The greater our success, the harder it may be to go on.? Mr Tiffany has a different worry. ?We can control the distribution of narcotics to a reasonable degree. We can control the violence.? But, with so many drugs pouring into the country and a popular culture that accepts them, ?we will reach the point where all we can do is to hold the line.?

Both in London and New York, the police rightly give priority to stopping the threats to public order and safety that drug-dealing can bring. Enforcement everywhere ought to have effects on the supply of drugs: it should drive up the price, reduce the competition and restrict the supply. But the increased efforts that governments have made to stem the flow do not appear to have raised the price, lowered the purity or discouraged the purchase or the use of drugs. That is true even in America, where policy has been concentrated on trying to reduce the availability of illegal drugs. This has been vastly expensive; it has sometimes corrupted the law-enforcement process; and it has damaged civil liberties and led to the imprisonment of hugely disproportionate numbers of non-whites. (…)"