The objective of this research was to estimate the
elasticity of the demand for cocaine and heroin with respect to the
price. Price elasticity is the percentage of change in the dependent
quantity corresponding to a one-percent change in price. The project
involved the development of an econometric model to determine price
elasticity, given that national- and city-level data on the
consumption of cocaine and heroin are insufficient or nonexistent. The
researchers circumvented this lack of data by partitioning the desired
elasticity into the product of two elasticities, involving a
measurable intermediate quantity whose relationship to the quantity of
consumption could be modeled and estimated by measurable
techniques. The intermediate quantity used for this project was the
fraction of arrestees testing positive for cocaine or heroin as
measured by the Drug Use Forecasting (DUF) System. From the Drug
Enforcement Administration's (DEA's) System to Retrieve Information
from Drug Evidence (STRIDE) data, expected purity was computed by
regressing on log quantity and dummy variables for location and
quarter. Price series were produced by finding the median standardized
price per expected pure gram for each location and quarter. Variables
for Part 1, National Data, include year, quarter, standardized prices
for a gram of cocaine and a gram of heroin, and expected purity of
cocaine and heroin. The Cities Data, Part 2, cover city, year,
quarter, number of observations used to compute the median price of
cocaine and heroin, standardized prices, and expected purity.

Over the last decade, illicit drugs have been
increasingly recognized as consumer goods. They are produced,
distributed, and retailed in markets that are not entirely unlike the
markets for licit goods. This research sought to answer questions such
as: (1) By how much will consumption change if prices increase? and
(2) If the price of one drug goes up, will users switch to another
drug? The answers to these questions have significant policy
implications. The primary objective of drug enforcement is to reduce
consumption and increase price, using interdiction and source country
control. However, the final step of translating the projected increase
in price into a reduction in consumption is never satisfactory,
because the elasticity of demand is not known. Better knowledge of the
extent of substitution would also aid policy planning, analysis, and
evaluation. Furthermore, interventions designed to affect price could
consider how the difference in the higher fraction of disposable
income absorbed by drug purchases for women and youth affect
demand. Therefore, the objective of this research was to estimate the
elasticity of the quantity of cocaine and heroin consumed with respect
to the price.

In general, price elasticity is the percentage of
change in the dependent quantity corresponding to a one-percent change
in price. The focus of this research was the development of an
econometric model to determine price elasticity corresponding to the
percentage of change in the quantity of cocaine and heroin consumed to
the change in the price of cocaine and heroin. The use of the standard
econometric model was not feasible, given that national- and
city-level data on consumption of cocaine and heroin are insufficient
or nonexistent. The researchers adapted the model by partitioning the
desired elasticity into the product of two elasticities, involving a
measurable intermediate quantity whose relationship to the quantity of
consumption could be modeled and estimated by measurable
techniques. The intermediate quantity used for this project was the
fraction of arrestees testing positive for cocaine or heroin as
measured by the Drug Use Forecasting System. From the STRIDE data,
cocaine and heroin purchase records were used if (1) they had a
positive purity, (2) the amount paid was greater than one dollar, and
(3) the total weight was between 0.05 and 40 grams. To partially
offset the situations in which cities' records were inadequate to
produce reliable quarterly price estimates, data were augmented with
records from adjoining cities. Gross outliers were eliminated with a
coarse filter based on price per gram, ignoring purity. A second step
eliminated observations whose standard price was more than five times
farther from the median standardized price for its location and
quarter than the median distance. Expected purity was then computed by
regressing on log quantity and dummy variables for location and
quarter. Price series were produced by finding the median standardized
price per expected pure gram of the remaining records for each
location and quarter, provided at least five such records existed. If
fewer than five records existed, the price was recorded as missing.

ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

- Standardized missing values.
- Checked for undocumented or out-of-range codes.

Variables for Part 1, National Data, include year,
quarter, standardized prices for a gram of cocaine and a gram of
heroin, and expected purity of cocaine and heroin. The Cities Data,
Part 2, cover city, year, quarter, number of observations used to
compute the median price of cocaine and heroin, standardized prices,
and expected purity.

AVAILABLE. This study is freely available to the general public.

The original collector of the data, ICPSR, and the relevant funding agency bear no
responsibility for use of the data or for interpretations or inferences based upon such uses.

Users are encouraged to obtain a copy of the technical
report on this project that describes in detail the model that was
developed the relationship between quantity consumed and the
fraction of arrestees testing positive. Further information about the
DUF program is available in various NIJ publications and can be obtained
from the National Criminal Justice Reference Service (NCJRS).