Paradox:
There has been a decreased in traffic deaths.
Yet, the alcohol consumption of those under 21 was the same as before the new regulation.

A: For the population as a whole, annual alcohol consumption is no lower now than it was in 1990
If consumption before and now is the same, how do we resolve the traffic-death decline? We are left in the dark here. OUT!

B: Alcohol consumption away from home, for example in bar and restaurants, is much lower among people under 21
than it was in 1990.
Most of the alcohol drinking is not away from home (or in bars)... Less drink and drive... This explains the paradox. BINGO!

C: The proportion of people under 21 who own a car is higher now than it was in 1990.
Okay? But why did the traffic-death decline after the new regulation. We are left in the dark here. OUT!

D: Alcohol consumption is lower among people under 21 than among adults in mist other age -groups
Still doesn't resolve the paradox.

E: Alcohol-related traffic deaths among people over 21 have increased slightly since 1990.
Still doesn't resolve the paradox about traffic-death decline vs. same alcohol consumption

The answer is B. Even if the alcohol consumption was the same, the traffic death decline was realized because of less drink and drive.