Understanding how price changes will impact on alcohol and other drug use is of great importance. Alcohol use is associated with a range of harms that mostly reflect a dose response model and there is strong evidence that increasing alcohol price is the most effective way of reducing alcohol consumption [1]. However, a possible unintended consequence of an increase in alcohol unit price is that people may switch from alcohol to other drugs or adopt more dangerous forms of consumption. This study investigates the effect of proposed alcohol price changes upon a university student sample, a demographic which has an established precedent of higher risk alcohol and drug use and a high sensitivity to economic restrictions. A recent study found that 50% of Australian university students were drinking at least once a week, 54% had used cannabis and 13% had used ecstasy at least once in their life, suggesting that substance use is a part of the student culture [2]; therefore the sample will provide a sensitive gauge of substance replacements patterns as a function of alcohol price increase.

Alcohol price and taxation

Numerous studies have reported reduced alcohol use [3], violence [4,5] and injury [6] when alcohol price is increased. A broad international body of research concludes that excise taxation appears to be the most successful alcohol policy in terms of cost-effectiveness, reductions in level of consumption, and overall social benefit [3]. Further, the tax-approach to reducing alcohol consumption generally yielded considerably larger effect sizes than any other prevention policies or programs [1]. While the relationship between price and consumption is clear, there is little evidence regarding how increase price affects consumption levels and which people are most likely to be affected by price change [7].

Economics of alcohol use

Previous research has predicted that the likelihood of drug use increases when there are ‘minimal’ constraints on drug availability and decreases when constraints are present [8]. Certainly, substance use decreases or changes with constraints on access (mainly price) to the substance [9-12]. As price is a major constraint, there is merit in investigating the relationship between substance use patterns among Australian university students as a result of hypothetical changes to alcohol price.

Previous research has demonstrated how demand for alcohol decreases as price increases [13]. Certainly, this effect was seen during the recent introduction of the ‘alcopops’ tax in Australia, a specific tax on pre mixed spirits or RTDs (Ready-To-Drinks), whereby there was an overall drop in alcohol consumption and people generally consumed less alcohol [14]. Chikritzhs et al. [15] reported that in the 3 months after the April 2008 tax increase, 91 million fewer standard drinks were sold as than in the same months in the previous year. However, substitution of one substance for another can also occur when the price of one substance increases [16]. This was also observed in regards to the introduction of the alcopopstax, where some drinkers substituted alcopops with beer or spirits [14]. While standard drinks sold as spirits and beer increased (48 million standard drinks), the increase was only 53% of the 91 million fewer RTD drinks sold [15].

There is also a small body of research exploring substitution with other drugs, although the link between changes in the price of alcohol and other substance use has not being directly made. In America, alcohol was made less accessible for most university students as a result of raising the legal drinking age to 21. It was found that while alcohol consumption decreased, use of cannabis increased at a one-for-one rate in high school seniors [17]. Another study in the UK found that as price of cocaine and amphetamine increased, alcohol was used as a substitute [16]. This suggests that people may substitute substances on the basis of price and availability.

This study will sample from a population known to participate in higher risk substance use and exhibit consumption patterns sensitive to economic change. This sample provides a sensitive indicator of whether unit increase in alcohol price will result in decreased alcohol consumption and possible substitution.

Participants

Five hundred and twelve students from Australian Universities responded to the online survey. The study was approved by Deakin University’s Human Research Ethics Committee. The selection criteria for participation were that the participants must be students at Australian universities and must have used drugs or alcohol at some stage of their life. The study was advertised on facebook, as well as on the Deakin University Portal, inviting people to ‘have their say on the alcopops tax’. The facebook advertisement was viewed on 4,763,550 occasions. Of those viewings, 1,561 (0.031%) individuals followed the link. There is no reference data to know whether this response rate is normal from facebook advertisements at the current time.

The majority of participants heard about the study on facebook (93%), were from Victorian universities (53%), and were living at home with their parents (56%). The age of participants ranged between 17 and 38 (M=20.31, SD=2.46) and participants were predominately female (66% female, 34% male). Twenty seven participants were excluded for not meeting selection criteria, leaving a sample of 485 participants.

Materials

To test for the impact of the increased alcohol price on alcohol and substance use, an online survey was used. The survey consisted of eight sections: 1) basic demographic information; 2) questions pertaining to substance (including alcohol) consumption patterns now and based upon a hypothetical unit price increase of alcohol; 3-4) participants’ desired effects from alcohol and ecstasy; 5-7) participant’s awareness of and/or experienced of effects associated with drinking alcohol, taking ecstasy, and using cannabis, and; 8) questions taken directly from the World Health Organisation-Alcohol, Smoking and Substance Involvement Screening Test (WHO-ASSIST) V3.0.

The questionnaire also contained questions adapted from the Desired Effects of Drinking Scale (DEoDS) [18]. The remaining 25 questions were constructed by the research team based on ideas from [13,19,20].

Questions related to alcohol price and consumption were expressed in terms of Australian ‘standard drinks’ and participants were supplied with reference information about how standard drinks related to the beverages they typically consume (i.e. a glass of beer).

Procedure

When participants clicked on a weblink to access the questionnaire, they were taken to an introductory page, which contained the plain language statement. By agreeing to the terms of the plain language statement, participants were then able to access the online questionnaire. On completion of the questionnaire, participants were given the option of obtaining a summary of the results.

General consumption trends

As shown in table 1, alcohol, followed by tobacco, then cannabis were found to be the most commonly used drugs in this survey. Amphetamine type substances (ATS), which include ecstasy, also ranked highly among participants with almost one third (28%) of participants having ever used them (Table 1). A nightclub was the most common venue attended on a night out by people who had ever used alcohol, cannabis or amphetamines (including ecstasy) (Table 2). White spirits (self mixed) ranked as the most common drink of choice by 61% of participants, while 49% reported alcopops as their preferred drink, followed by 40% reporting beer. When asked about what they normally drink on a night out, 55% of participants reported white spirits (self mixed), 43% reported beer and 34% reported alcopops (Table 3).

Table 1: Self reported drug use (N=512).

Table 2: Usual venue on a typical night out by drug type ever used.

Table 3: Participant’s alcoholic drink of choice and drink typically purchased on a night out.

Desired intoxication level

Of the 485 participants in the current study, 54 (11%) report that they drink to get drunk ‘always’ and 75 (16%) report that they ‘never’ drink to get drunk. As shown in table 4, participants reporting that they drink to get drunk ‘always’ are more likely to report illicit substance use in the past 3 months. Substance use (ever) for participants reporting that they drink to get drunk ‘always’ is significantly different from substance use (ever) for the rest of the participants, χ²(7, N=484)=14.35, p<.05. However, substance use (past 3 months) for participants reporting that they drink to get drunk ‘always’ is not significantly different from substance use (past 3 months) for the rest of the participants(χ²(7, N=484)=8.014, p>.05).

Table 4: Desired intoxication level of participants by substance use (past 3 months).

Participant price choices

As shown in figure 1, while almost 100% of participants reported that they would pay up to $4 for a drink when drinking four or more drinks, less than 25% of participants were willing to pay $10 or more for a drink.

Figure 1: Cumulative percentage of all participants reporting viability of substance use options at proposed per drink price points (dashed line indicates significant cut-off points).

Almost half (49%) of all participants reported that they would never opt to use cannabis at any of the 20 proposed price points. The percentage of participants who reported cannabis as a viable option increased when the price per drink rose above $10. Of those participants that would consider cannabis a viable option, 83% (N=203) had used cannabis before, while 17% (N=41) had never used cannabis before but would consider it a viable option as drink prices rose to $20.

Over half (58%) of all participants, reported that they would never use ecstasy. The percentage of participants who reported ecstasy as a viable option increased sharply when the price per drink rose above $10. Of those participants that would consider ecstasy a viable option, 62% (N=124) had used ecstasy before, and only 15% of the sample who had never used ecstasy before would consider it a viable option as drink prices rose to $20.

Participants who self-report that they drink to get drunk ‘always’ were significantly more likely to consider cannabis and ecstasy as viable options as the price of alcohol increases (χ²(1, N=483)=9.39, p<.001 and χ²(1, N=484)=20.50, p<.001 respectively). Seventy percent of the participants that report drinking to get drunk ‘always’ would consider cannabis and ecstasy a viable option as drink prices increase, compared to 48% of the rest of the participants for cannabis and 38% for ecstasy. Participants who had ever used amphetamines were willing to consider substituting ecstasy for alcohol at a lower mean price per standard drink ($9.66) than those who had never tried ecstasy ($12.82, t (210)=-4.68, p<0.001). Similarly, participants who had ever used cannabis were willing to consider substituting cannabis for alcohol at a lower mean price per standard drink ($10.31) than those who had never tried cannabis ($13.17, t (255)=-2.708, p<0.01).

As expected, higher risk substance use is consistent with previous samples of university students. Over half (58%) of the sample had used any drug, 53% had tried cannabis and 28% had tried amphetamines in their lifetime, which reflect findings of previous international and Australian studies [2,21,22]. These heavy substance use patterns are also reflected in alcohol consumption figures where 98% of the sample had consumed alcohol within the past three months.

Alcohol consumption and price

Consistent with Babor et al. conclusions [3], unit price and proposed alcohol consumption are negatively related. The results clearly show a downward sloping demand curve, with demand for alcohol decreasing as price rises. These results are consistent with a number of other studies that also found increasing drink prices leads to reduced consumption in university students [13,23]. Figure 1 shows a consistent decline in likely consumption as drink price is raised from $3 to $10 per drink. Further, they were virtually no participants who reported that they were engaged in a session of four or more drinks (the current Australian National Health and Medical Research Council guideline for episodic drinking) once the price went above $14 per standard drink. This suggests that the price of around $14 per standard drink is likely to result in drinking levels consistent with public health guidelines. In this context, it is worth also considering that bottle shops currently sell heavy beer on special at around $1 per standard drink, dark spirit RTDs at around $2 per standard drink and cheap vodka at around $1.25 per standard drink (http://www.danmurphys.com.au).

Alcohol price and other drug use

While the majority of the sample (58%) would never consider using ecstasy or cannabis (49%), there was a positive relationship between unit price for alcoholic drinks and likely ecstasy and cannabis use. As price per drink rose, so too did the number of participants who would consider ecstasy and cannabis as a viable option. People who had previously used ecstasy or cannabis were likely to substitute for alcohol at a significantly lower price (around $10 per standard drink) compared to those who had never used either drug (around $13 per standard drink). For those that would consider switching to other drugs, the price of alcohol would need to exceed $14 per standard drink before one third would consider ecstasy a viable alternative and 42% would consider cannabis a viable alternative (Figure 1). This is especially significant given that substitution and price elasticity effects are exaggerated due to the higher financial restrictions associated with student life, and those base levels of consumption and acceptance of illicit substance use amongst this population are higher. Therefore, to see significant changes in illicit drug use related to alcohol price, the price per standard drink would have to be between 10 and 14 times more expensive than current sale prices from bottle shops.

This study relied on self-report data, making it possible for participants to distort their answers and answer untruthfully. The sample for this study was largely drawn (85%) online via facebook, and the recruitment method targeted participants who were interested enough to ‘have their say’ regarding the alcopop tax. This sampling method was effective in recruiting a participant demographic which traditionally has higher risk substance use patterns than the general population. While this has resulted in findings which are sensitive to economic change and which highlight the economic choices associated with substance use, a much broader recruitment method would be preferable in future studies. As this was the first study of its kind in Australia, it highlights the need for further research to be conducted in this area, in an attempt to address these limitations and replicate the findings. Future research could also focus on the impact of variation in price of illicit substances upon consumer viability. Finally, females were over-represented in the sample and future studies should attempt to achieve similar numbers of male and female participants.

The findings from this study indicate that increasing the price of alcohol will decrease consumption, with the greatest effect noticeable between the $9-14 per standard drink. Even amongst a demographic that has a precedent of higher risk illicit substance use, the majority of participants would not consider substituting ecstasy or cannabis for alcohol. A substantial minority of young people report that they would consider substituting alcohol with other drugs once price per standard alcoholic drink reached $14. Given heavy beer can be purchased for as little as one dollar per standard drink in Australia at the moment, this finding suggests there is substantial room to increase the price of alcohol before substitution becomes an issue, even more so amongst the general population. However, further study is required in a more representative population, as substance use and financial constraints are skewed in the current sample.

The authors declare that there are no competing interests.