Wet & Dry Counties in Texas, Tennessee, and Kentucky

Of the many controversies surrounding the consumption of alcohol, one of the most stringent counter-measures to ensure safety of consumers is the outright banning of the sale of alcoholic beverages.  This phenomenon of “dry” areas of the nation has been prevalent since the Puritan settlements of the American colonial period during the 18th century.  The most widespread ban on alcohol sale in US history occurred with the passing of the 18th amendment in 1919 commonly known as the Prohibition Act.  It banned the manufacture, sale, or transportation of alcohol.  After the repeal of this amendment in 1933 with the passing of the 21st amendment, several regions of the nation chose to maintain the regulations in place under Prohibition.  Federal laws dictate that states themselves cannot directly regulate this specific aspect of commerce, but local governments such as counties, municipalities, and townships can pass laws to that effect.  Nearly 80 years since the repeal of Prohibition, over 500 cities and townships remain dry in some form to this day.

Since these regulations are enacted differently in each individual county that chooses to remain dry, there is no uniform system for classifying counties as either wet or dry.  Many states have adopted a system which categorizes counties as either wet, dry, or moist, with moist counties having some form of regulation surrounding the sale of alcohol, but not making it completely illegal.

Many different motivations persist that encourage a county to stay dry, many of them religious-based reasons.  However, many ordinances cite the safety of residents and consumers as a major incentive for prohibiting the sale of alcohol in a county.  The motive behind these ordinances is that by limiting access to alcohol, citizens will pose less risk to each other when choosing to drink.  Opponents of these laws however have reasoned that residents of a dry county have to drive further to get alcohol, thereby potentially increasing the number of instances of driving under the influence (DUI) and the resulting fatalities which can occur.

This study examines the wet and dry counties of Texas, Tennessee, and Kentucky for the effects of these regulations on the number of DUI-related fatalities observed in a county.  As there is no standardized categorization of wet and dry counties between different states, it was assumed that for the purpose of this study, counties will be considered dry if they prohibit the off-premises sale of alcohol (retail sale for consumption off the premises of the store).  Counties will be considered wet if they allow off-premises sales.  All county wet/dry information was taken from the most recent elections and citizen decisions as of November 2012.  This process effectively classified the following number of counties in each state as wet or dry:

State Wet Counties Dry Counties Total
Kentucky 63 57 120
Tennessee 59 36 95
Texas 232 22 254
Total 354 115 469



Once counties were classified as either wet or dry, population information was compiled for each individual county in the studied states from the 2010 census.  Cities greater than 10,000 people were then displayed as a new layer on the map to highlight where there are population centers in these areas.



As might be expected, a large number of dry counties are in rural areas.  All of the dry counties in Texas have cities with less than 10,000 residents and therefore none of the cities of this layer lie in dry counties.  Tennessee and Kentucky both have more evenly dispersed populations with several cities greater than 10,000 residents in dry counties.  These initial findings of populations in dry counties sets the stage for the later findings of this study where smaller populations observe a higher proportion of the ill effects of alcohol consumption.

Below are population statistics for the dry counties of Texas


             Yearly information on alcohol related vehicular fatalities for each state, organized by county is maintained by the US Department of Transportation .  After cleaning this data into a compatible format, DUI related vehicular fatalities were incorporated into the county maps of the study regions.  As would be expected, areas with a larger population have the highest number of DUI vehicular fatalities overall.  For the Texas map below, the areas of Houston, Dallas/Ft. Worth, Austin, and El Paso have distinctly higher numbers of fatalities than in other regions.


However, when the number of fatalities was normalized with the population of each county, the layout of the map changed significantly.  The observed ratio of fatalities to population total in a county was most pronounced in counties with smaller populations.


In Kentucky and Tennessee, these observations were not as distinct, since the counties in these two states are overall more evenly populated than those in Texas.


           In the less populated areas of Texas, the fatalities observed per population were much more pronounced.  Because of the extremely small population, some Texas counties where only one fatality occurred displayed fatality rates of greater than 1 fatality per 1000 residents.  As there were no extreme outliers in the counties of Tennessee or Kentucky, the bulk of the findings of this study concentrate on the counties of Texas.  The highest ratio was observed in Kenedy county Texas, where 1 fatality occurred per 128 citizens.  These extremely high ratios can be attributed to the low population numbers found in these areas, which provide a smaller base of residents to distribute the fatalities among when calculating the ratio.

The table below displays information for counties in Texas which recorded greater than 1 fatality per 1000 residents.  The Fatal_Pop field shows this ratio.


            The high fatalities per population ratio counties are not exclusively dry or wet.  For the 16 Texas counties observing a ratio greater than 1 in 1000, 4 of these counties are dry and 12 are wet.  The map below displays those 16 regions and the corresponding colors for wet and dry counties.


           It is interesting to note that though 25% of the high fatality ratio counties are dry, fewer than 9% of the counties in Texas are considered dry by the assumptions of this study.  However, 82% of the dry counties in Texas have fewer than 10,000 people occupying them and the highest population density of a dry county is 37 people per square mile.  Therefore, these counties generally have a higher ratio than other more densely populated regions, it would seem, because any DUI related fatality in such a small community would skew this ratio disproportionately.

The counties in Tennessee and Kentucky observed lower ratios than the Texas counties.  In fact, not one county in either state had a fatality-to-population ratio higher than 1 in 1000.   This is partially due to the more evenly distributed population in these states.  Though there are still significant rural areas in each state, they may not be nearly as rural as the counties in Texas.  Counties in Texas appear to be either very densely populated in urban regions or extremely sparsely populated in the rural areas.  Even though both Tennessee and Kentucky have larger numbers of dry counties than Texas, (Kentucky has more than twice those in Texas) their dry counties are not nearly as rural, and as a result, have lower fatality ratios.  This finding is particularly interesting as it seems to contrast the trends that the Texas counties indicate about dry counties.  A scatter plot of the county ratios each state was compiled to illustrate the trends observed in each specific state for fatality ratios compared to county population.

TXFatalPopGraph KYFatalitiesPopGraph TNFatalitiesPopGraph

            In all three states, the trends shown in the scatter plot diagrams indicate that the large population centers of each state, those cities over 500,000 people, seem to reach a leveling out ratio of 1 in 10,000 people.  The Texas graph is significantly distorted compared to Tennessee and Kentucky because it has a significantly wider scale spanning from 0 to 7.38, whereas the scale for Kentucky and Tennessee do not span more than 0 to 1.0.  After eliminating the 16 outlier counties with fatality ratios over 1 in 1000 residents, the graph much resembles the findings for Tennessee and Kentucky.


            Overall, these findings indicate that the existence of dry counties does not necessarily jeopardize or ensure safety for the residents in these counties.  The highest incidence of DUI related fatalities normalized with county populations all occurred in very rural areas of Texas, regardless of the counties being considered wet or dry.  Another telling indicator is the fact that, though there are significantly more dry counties in both Kentucky and Tennessee, and the combined populations of these two states are less than that of Texas, they still exhibited very low fatality-population ratios in comparison.  The high DUI fatality ratios are more an indication of low population areas than they are of highway danger zones.  These results did not correlate with the counties being wet or dry, but did denote that counties with high ratios are typically rural and overall outliers in comparison to the entire study region.  If more conclusive results are to be sought, this study should be expanded to encompass a national scope of wet and dry counties.

This entry was posted in 2012. Bookmark the permalink.

One Response to Wet & Dry Counties in Texas, Tennessee, and Kentucky

  1. The banning of alcohol is a welcome development. DUI fatalities is quite alarming. Texas drug related crimes is high because of the population involved.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s