Analyzing the effect of drug prohibition on drug use is difficult because the data necessary for such an investigation are limited and low quality (National Research Council 2001; Horowitz 2001).Read more at location 440
reliable data on drug use do not exist before the onset of U.S. drug prohibition in 1914,Read more at location 443
One useful piece of evidence, however, is the U.S. experience with prohibition of alcohol, which occurred from 1920 through 1933.Read more at location 445
In addition, cirrhosis death rates constitute a good proxy for the consumption of alcohol,Read more at location 448
The bottom line is that Prohibition appears to have reduced cirrhosis death rates by 10–20 percent. This is not a trivial effect, but it is far smaller than suggested by many advocates of prohibition.Read more at location 454
Merck & Co (1992: 890) states that “in general, a linear correlation exists between the intensity of alcohol abuse in terms of duration and dose and the development of liver disease.”Read more at location 459
One issue is that cirrhosis is probably a better proxy for heavy alcohol consumption than for moderate or light alcohol consumption.Read more at location 461
A second problem is that cirrhosis typically develops only after years of alcohol consumption;Read more at location 463
The data show that cirrhosis was substantially lower after the onset of national prohibition (in January 1920) than it had been in most of the pre-Prohibition period. The death rate declined from 12–14 deaths per year per 100,000 population during the 1910–1915 period to 7–7.5 during the Prohibition period. This is the fact typically cited as indicating Prohibition caused a substantial decline in alcohol consumption.Read more at location 469
the cirrhosis death rate was already at its minimum level when Prohibition took effect in January 1920.Read more at location 477
so the low level in the 1920s is not by itself evidence of Prohibition’s efficacy in reducing cirrhosis.Read more at location 480
data do not show a sudden or dramatic increase in cirrhosis after repeal in 1934. Instead, cirrhosis increases only gradually over several decades, and cirrhosis declines substantially starting around 1970, well after Prohibition ended.Read more at location 484
A possible response to this last argument is that even if alcohol consumption increased substantially after repeal, it would have taken many years for this increase to raise the death rate from cirrhosis.Read more at location 486
This hypothesis is unconvincing, however, for several reasons. First, it suggests that even if cirrhosis did not jump in 1934, it should have jumped, say, ten years later in (lagged) response to the jump in alcohol consumption. Such a jump is not evident in Figure 3.1. Second, this hypothesis is inconsistent with data on admittances to hospitals for alcohol psychosis and on deaths due to alcoholism (Miron and Zwiebel 1991); these two series are likely related to alcohol consumption with shorter lags than cirrhosis. Third, the hypothesis implies that during the first years after repeal, the amount of cirrhosis relative to alcohol consumption should have been unusually low; setting aside the first year or two after repeal, when official statistics almost certainly understate alcohol consumption, this is not apparent in the data.Read more at location 487
One factor to consider is state-level prohibitions of alcohol, which were adopted at an increasing rate during the 1910–1920 period.Read more at location 517
Although the number of states with prohibition was large, these states were predominantly rural, low population states.Read more at location 521
The role of state prohibition is also not compelling because the laws in many states were weak; in particular, they fell far short of bone-dry prohibitionRead more at location 526
For example, Alabama allowed any citizen to import two quarts of distilled spirits or two gallons of wine or five gallons of beer every fifteen days.Read more at location 528
The figures show that the most dramatic declines in cirrhosis occurred in states that were wet throughout the pre-1920 period, and these states included several of the most populous states (e.g., New York, Pennsylvania, New Jersey).Read more at location 536
An additional reason to question the role of state prohibitions is that adoption of alcohol prohibition is not an exogenous event, imposed on a state by forces outside its control.Read more at location 553
States in which per capita alcohol consumption was relatively low might have been more likely to adopt dry laws. Similarly, states in which consumption was declining for other reasons (e.g., changing demographics) might also have faced less opposition to dry laws.Read more at location 555
A second possible explanation for the pre-1920 decline in cirrhosis is federal regulation of alcohol.Read more at location 559
These restrictions did not prevent consumption of imports or existing stocks, and the budget for enforcement was essentially zero.Read more at location 569
Further, most of these restrictions did not take effect until 1917 or later, while cirrhosis began declining as early as 1908Read more at location 570
a number of other factors likely played a more direct role in the declines from 1917 to 1920, including a drastic reduction in immigration that took place during the earlier part of the decade, a major increase in alcohol tax rates that occurred in 1916–1917, World War I, and the worldwide flu epidemic of 1918, which killed tens of millions (Kolata 1999).Read more at location 571
A complete analysis, however, must account for a range of factors that potentially influence alcohol consumption, such as income, demographics, and alcohol tax rates. Dills and Miron (2003) present a detailed statistical analysis that accounts for these factors, concluding that national prohibition reduced cirrhosis by 10 to 20 percent.Read more at location 577
This conclusion is surprising, since standard accounts suggest alcohol prices rose substantially during Prohibition, perhaps by several hundred percent on average (Warburton 1932; Fisher 1928). Thus, since available evidence suggests alcohol consumption is responsive to price (Leung and Phelps 1993), alcohol consumption should have declined dramatically. One possible reconciliation is that the relevant price elasticity is in fact quite low. The proxy for alcohol consumption considered here, cirrhosis, is plausibly a better measure of heavy consumption than of moderate consumption.Read more at location 579
A second possibility is that Prohibition created a forbidden fruit effect, thereby shifting preferences for alcohol and partially offsetting the depressing effect on demand of higher prices.Read more at location 586
a third possibility is that the standard accounts of alcohol prices during Prohibition overstate the increase in price.Read more at location 590
the data presented by Warburton and Fisher overstate the increase in the relative price of alcohol.Read more at location 593
since other alleged benefits of prohibition, such as reduced crime, improved productivity, or better health, depend on the decrease in consumption, these benefits are likely modest as well.Read more at location 604
alcohol prohibition and current drug prohibition are not comparable, since the strictness of the law and the degree of enforcement have been greater under drug prohibition than under alcohol prohibition.Read more at location 606
the analysis in Miron (2003a) suggests that despite the enormous level of resources devoted to enforcement, drug prohibition has not raised drug prices to nearly the degree suggested in most accounts.Read more at location 612
Previous analyses have suggested that prohibition makes drugs ten, twenty, or even hundreds of times more expensive than they would be if legal.9 Much of this analysis, however, simply notes that the raw materials from which drugs are produced sell at “low” prices in producer countries while the finished products sell at “high” prices in consumer countries, implicitly attributing the entire “markup” to prohibition. Such an analysis, however, does not account for the storage, transportation, distribution, and retailing costsRead more at location 613
the farmgate-to-retail “markups” on many legal goods (such as coffee, chocolate, tea, or beer) are similar to or greater than the markups on cocaine and heroin;Read more at location 620
calculations imply that the black market price of cocaine is two to four times and the price of heroin six to nineteen times the legalized price.Read more at location 623
over the past twenty years, enforcement of drug prohibition has increased substantially, but real, purity-adjusted drug prices have generally declined (Basov, Jacobson, and Miron 2001).Read more at location 626
price of cocaine falling from over $450 per pure gram in 1981 to roughly $100 per pure gram in 1996.Read more at location 628
DiNardo (1993) finds no evidence that enforcement, as measured by cocaine seizures, raised cocaine pricesRead more at location 629
Yuan and Caulkins (1998) find that a greater number of drug seizures is associated with lower black market prices of cocaine and heroin. Basov, Jacobson, and Miron (2001) show that despite the enormous increase in prohibition enforcement that has occurred over the past twenty-five years, drug use appears little different now than at the beginningRead more at location 631
There is thus little evidence that enforcement under current drug prohibition has raised drug prices or decreased drug consumption to a substantial degree. More generally, several kinds of evidence fail to indicate a substantial impact of drug prohibition on drug consumption.Read more at location 637
existence of such substantial fluctuations in the absence of prohibition suggests many factors affect opiate consumption, not just government policy. These factors include demographics, per capita income, wars, and the like.Read more at location 643
A different kind of evidence comes from the experience of U.S. states that decriminalized marijuana at various points during the 1970s.Read more at location 648
The evidence provided by decriminalization is potentially weak; changes in the law sometimes ratify ex post what has already taken place,Read more at location 651
Nevertheless, existing evidence provides little indication that marijuana decriminalization was accompanied by increased marijuana use.Read more at location 655
One additional piece of evidence comes from comparing drug use rates between the United States and other rich countries such as those in Western Europe, Japan, or Australia.Read more at location 656
other countries have prohibition laws similar in broad structure to those in the United States, especially for harder drugs such as cocaine and heroin. The degree to which these countries enforce their prohibition regimes, however, is markedly less (Miron 2001b). Thus, if prohibition is an effective method of reducing drug use, these countries should have use rates noticeably higher than in the United States.Read more at location 659
As shown in Table 3.2, however, there is no evidence these countries have higher drug use rates; indeed the U.S. rate frequently exceeds that in most other countries. This evidence is only suggestive, since it does not control for other factors.Read more at location 662
If one credits Prohibition for the entire fall in cirrhosis from, say, 1915 to 1925, this suggests Prohibition reduced alcohol consumption by roughly 50 percent.Read more at location 666
