Ipsos Thinks

Youth gone wild / youth gone mild

In brief Gen Z and rebellion

  • Recent years have seen a generational decline in a number of behaviours associated with rebellious youth. When compared to Millennials at the same age, Generation Z drink less, engage in fewer sexual activities and commit less crime.
  • Drinking culture among teenagers has weakened across Europe and the US. In the UK, Gen Z 13-15 year olds are nearly half as likely to have even tried alcohol as Millennials at that age in 2004.
  • In the UK at least, this could be down to a real shift in risk perception among teenagers – more than 70% think binge drinking is very risky and 28% think having any alcoholic drink is very risky.
  • However, with the exception of alcohol consumption, this decline does not seem to be due to teenagers changing their risk perception. In fact, they are less likely to think drug use and unprotected sex are risky than the previous generation of young people.
  • Generation Z are also less likely to commit crime or get into fights. In Germany, the number of arrests of 12-16 year olds dropped by 40% between 2006 and 2015. And in the US, physical fights in high schools are far less common than when Millennials were in school in 1999.
  • But generational drug taking is more varied from country to country and drug to drug. In the US, a 12-percentage point drop in high schoolers trying marijuana is juxtaposed by very stable trends in the UK.
  • And although teenagers in some countries may be engaging in less sexual activity, there’s no real change in engaging in unprotected sex.


Are you scared yet?

On the surface – this milder form of youth is very encouraging. Typically bad habits, such as drinking, smoking, taking drugs and engaging in sexual activity, are considered more risky for children and young people (ignoring any legal issues for a second). This is partly because teenagers are at such a crucial developmental stage, but also that habits ingrained from an early age can be harder to shake off in adulthood. With Gen Z less likely to engage in these types of habits, there could be many positive implications for their future health, both mental and physical.

But whether this aversion to risk is permanent is not so clear. There is speculation that decreases in certain types of behaviour could be displacement as a result of increasingly digital lives. Young people get more of their thrills online, and are just too distracted to get up to real-world mischief.

People who study risk, such as University of Cambridge professor David Spiegelhalter, distinguish between hazard, which is the potential for harm, and risk, which is the probability of that adverse outcome actually happening.14 And, of course, while the hazards are different, the risks of online behaviour are real, and less likely to be monitored. Official crime statistics are only now catching up with cybercrime – and digital antisocial behaviour and the extent of inappropriate sexual activity online are not really measured at all. So Gen Z may be just as badly behaved – but in a different way.

There’s also evidence that potentially it is more to do with the increase in number and effectiveness of barriers which pre-empt and prevent bad behaviour. There’s no doubt that there have been general shifts in attitudes, behaviours and legal practices among western countries, which mean there has been a general population-wide decline in these behaviours. Measures massively vary across countries, but many have seen changes in policing methods and legal constraints which could have impacted on population-wide behaviours, as well as on young people – for example, changes in methods and preventative actions for youth crime, sanctions for drunken behaviour or for possessing and dealing drugs. Declines in these behaviours among young people are also linked to alterations in schooling methods – particularly among Scandinavian countries.

One thing that new data from our study does seem to suggest is that it’s not about a sudden growth in fear of these behaviours (alcohol excepted). This is counter to what one might assume, and intriguing – it is not a result of communications getting the message across better, or a general cultural shift towards caution. There does not appear to be a shift in young people’s internal risk calculus, which suggests that the changing cultural, technological and legal environment is key. This means that the level of support must be kept up if we want young people of the future to continue this trend.


Seedlip, the world’s first non-alcoholic distilled spirit, was developed by tee-totaller Ben Branson who was sick of only ever ordering tonic water in bars.

Made by soaking herbs and botanicals in neutral grain spirit and distilling to concentrate the flavours, the final product was developed over two years of tinkering in his kitchen.

By 2015, Seedlip had won a listing in Selfridges followed by Fortnum & Mason and Ocado. Then, in 2016, drinks multinational Diageo took a 20 per cent stake in the company through its investment arm Distill Ventures.

A wise investment. With year-on-year growth of over 1,000%,15 Seedlip is now stocked in around 400 bars and restaurants in the UK and has made headway in European cities, the US and Australia.

A future less fuzzy

Drinking alcohol is simply not as widespread among teenagers today, and it’s set to become a habit. With just over one in four of 16-24 year olds teetotal – about eight percentage points more than ten years ago – it’s likely this cultural shift will change the drinks industry.16

The low- and no-alcohol drinks market is booming. Top-quality ‘mocktails’ feature in an increasing number of bar menus. Fuelled by a tranche of new products, such as Heineken 0.0, Budweiser Prohibition and Guinness Open Gate Pure Brew lager, sales of low- and no-alcohol beer grew by 20.5% in the UK in 2017.17 Meanwhile in 2016 Diageo, the multinational alcoholic drinks company, made the pivotal decision to invest in the first non-alcoholic distilled ‘spirit’ Seedlip.

Forward-thinking food and drink manufacturers will follow suit. To tap into this teetotal market they need to think just as much about their non-alcoholic offering as they are about their alcoholic one.


The decline of alcohol

Alcohol consumption across more established markets has faced a relatively consistent pattern of decline over the past 15 to 20 years.18 People in the US and much of Europe are drinking less alcohol generally, but the freefall in adolescent drinking rates has fuelled a generational shift. Millennials drank less than their older counterparts when they were young – and now Generation Z are drinking even less.

And this international decline is despite the very different contexts, drinking cultures and norms in these countries. For example, the UK has seen one of the sharpest and most surprising shifts in drinking culture. Generation Z teenagers (specifically 13-15 year olds) are half as likely as Millennials at an equivalent point to have even tried alcohol. In 2016, just over a third (36%) of 13-15 year olds had experienced alcohol.19 Yet in 2000, nearly three quarters (72%) of 13-15 year olds had had at least one proper alcoholic drink.

This is a monumental change in children’s first relationship with alcohol and is mirrored, although less dramatically, in the US where 60% of high school students (aged 14-18) had tried an alcoholic drink in 2017 – much lower than the 81% of Millennial high school students in 1999.20 In Australia, cohort analysis pinpoints a generational drop in both drinking participation and volume among younger Millennials and older Generation Z, compared to older cohorts.21

No nose for crime

A strikingly similar pattern of youth crime rates and violence can be seen across different western countries. Recorded youth crime rates have declined in many different countries since Generation Z have hit adolescence.22 In 25 EU countries, the number of detained young people dropped by 48% in the period between 2008 and 2015.23 In Germany, for example, the number of arrests of 12-16 year olds dropped by 40% between 2006 and 2015.24

In England and Wales over the past decade, the number of recorded crimes committed by 10-17 year olds is three times less.25 If we look at the long-term trend (in the graph below) the sharp drop in youth offences from the late 2000s into 2010s occurred within the younger members of the Millennial generation and into Generation Z. Then, if we compare the cohorts like-for-like, in the year 2001-02, there were approximately 262,000 offences committed by Millennials aged 10-17. By 2016-17, the number of offences committed by 10-17 year olds (Generation Z) had dropped to c.73,000.

As with alcohol consumption, declining crime rates are not occurring among young people in isolation – there are wider patterns of decreasing crime rates in these countries. However, the drop is much starker for young people.

Young people also seem to be behaving less violently. For example, in the US, there has been a drop in the number of teenagers involved in physical fights in school. In 2015 (when Generation Z were up to 19 years old), 23% of high schoolers had been in a physical fight that year (17% of females and 28% of males). Whereas in 1999 (when Millennials were aged up to 19 years old), 36% of high schoolers had been in a physical fight (27% females and 44% males).

Keeping off the grass

However, drug use has seen less of a uniform decline among teens. In some countries there has been a decline, but in others, rates are steadier.

In the US, where a cultural shift on marijuana in particular has led to a number of states legalising cannabis, adult use rates have remained broadly steady. But Generation Z high school students are less likely to try marijuana than Millennial students were. The Youth Risk Behaviour Study among high school students records a 12 percentage point decline between cohorts. In 2015, 39% of Generation Z high schoolers had tried marijuana, compared to 1999, when nearly half (47%) of Millennial teenagers had given it a try.26

However in Europe, where measures of cannabis use are lower than those found in the United States, rates of use among teenagers remained stable – even showing a slight increase over the past 20 years among 15 and 16 year olds (5% in 1995 to 8% in 2015).27

S-exiled: less sex, more sexting

Sex among teens is difficult to measure – partly because of over and underreporting – but also partly because there are so many different ways to define sexual behaviour. With the data available, it certainly seems that at least some types of sexual activity are on the decline.

In the US, surveys among high schoolers have shown a slight drop in sexual activity between Gen Z and Millennials. Generation Z high schoolers are less likely than Millennial high schoolers to say they have had sex in the three months prior to being surveyed. In 2015, 30% of Generation Z in high school (up to 19 years old) were sexually active (with the caveat we are relying on self-reporting). In 1999, when Millennials were at an equivalent point, 36% of high schoolers were sexually active.

In our ‘Millennial Myths’ report, we found that Millennials were having sex less often and with fewer people compared to Generation X at a similar age. Research has now shown this generational decline in sexual activity and dating extends into Generation Z.28

However, there are some particularly risky types of sexual behaviour that are either just as prevalent or even potentially on the rise among Gen Z teenagers. Unprotected sex in the US is just as prevalent among high schoolers as before.29 In 2015, 14% of sexually active high schoolers said they had not used protection when they had last had sex – the same proportion of Millennials who were asked the same question in high school in 1999 (15%).

Meanwhile, ‘sexting’ is a new sexual activity among Generation Z, something that wasn’t nearly as available for previous generations of young people. And, although prevalence is difficult to measure, there is evidence that sexting among young people is gaining more traction and can be a predictor of sexual activity, as well as being associated with other risky behaviour.30

Risk perception is NOT the driver

Patterns of generational decline in risky behaviours is generally considered widespread enough to not simply be coincidence.31 So what is it about Generation Z which makes them a milder brand of youth than their predecessors?

One theory is that ‘period effects’ (see Introduction) around social attitudes and acceptability of certain behaviours have rubbed off on children growing up in the late ‘90s and ‘00s. Parents, teachers, media and wider society waxing lyrical about the dangers of certain types of behaviours has rubbed off on adolescent perception of risk.

However, trend data from UK secondary school children indicates otherwise. Asking 11-15 year olds how risky they consider certain behaviours to be shows Gen Z teenagers today are actually less worried on most measures than Millennials were at an equivalent point in 2004. They’re less likely to think various types of drugs are very risky. For example, seven in ten (72%) adolescents think smoking cannabis is very risky, compared to 84% of Millennial teenagers in 2004. They’re also less likely to think unprotected sex is risky (57% compared to 63%), less worried about walking alone at night in strange areas (57% think it is very risky compared to 67%) and less likely to see smoking as high-risk (70% compared to 76% of Millennials in 2004).

The only exception to this rule is binge drinking. Generation Z are much more likely to think this is high-risk than Millennials did – seven in ten (70%) think this is very risky compared to just 56% of Millennials at an equivalent point in 2004.

This means that the stark generational behaviour change is not due to a concomitant change in how risky young people view other types of behaviours.

Arguably, a reduction in these types of bad behaviour may be less to do with shifts on risk perception and more to do with changing social norms. People are more likely to do things if they think others are doing them too – teenagers today are simply coming into personal contact less frequently with these types of behaviour. And in turn, they are less personally affected by it. This is good, because social norms become self-reinforcing. But this lack of contact can be double-edged, because if people are less likely to see the negative impact of the behaviours this may make them more complacent.


As we’ll see throughout this report, there is no one clear shift on risky behaviour, and not all of Gen Z are behaving in a particular new way on drugs and sex.

But the balance has definitely tipped towards milder rather than wilder youth. And this is a continuation of trends seen in Millennials, not a sudden switch. It has been embedding over many years.

But it would be a mistake to assume this means that it’s guaranteed into the future. As our new data on perceptions of risk shows, this is not a change in how risks are viewed by the young. It’s not due to a shift in young people’s internal risk calculus, but instead suggests that changing cultural, technological and legal environments are far more important. This is not a guaranteed downward trend and we cannot sit back and hope it will continue. Continued support and intervention will remain key to ensure that future cohorts of young people follow this downward trend.

But it also offers up a plethora of new opportunities for brands and other institutions to engage with young people in different ways. Thrills will be found elsewhere, and products need to be updated for this new outlook.

The Perennials: The Future of Ageing

This report looks at our ageing societies and their challenges and opportunities. The media have obsessed about ‘Millennials’ as disruptors, but in fact ageing – which will affect every generation – is having even larger effects. Rising life expectancy and falling birth rates mean populations around the world are getting older. It is portrayed as a ’narrative of decline’ – not a time of opportunity and change.