National Service, another failed idea that never dies, is in the news again.

It would be tempting to dismiss the proposals that are currently being discussed as a short-lived election gimmick, and there is a good chance that that is precisely what they are. However, like most terrible policies, the idea enjoys a fair amount of popular support. According to a recent YouGov poll, 47% of the public are in favour of it (including about a third of young and middle-aged people), while 45% are opposed. A similar poll by JLP finds 42% in support, but only 34% opposed. So while we should probably not pay too much attention to the specifics of the current election circus, we cannot entirely dismiss the broader concept of National Service.

Proponents of National Service usually describe an idealised, hypothetical version of it. They do not pay much attention to how National Service actually turns out in practice. This is a surprising omission. Even though Britain has not had National Service for a long time, many other OECD countries only abolished it much more recently, and some still have at least a watered-down form of it to this day.

Whether National Service “works” or not depends on what we think it is supposed to achieve, which is a matter of political preferences and value judgements. But it is safe to say that National Service imposes non-trivial economic costs, which its supporters, even if they are not especially interested in economic outcomes, should not brush aside.

In 2006, the Bonn-based Institute for the Study of Labor (IZA) published a study called “Military Draft and Economic Growth in OECD Countries”, which summarises the existing literature on the subject, and adds an estimate of its own.

Mandatory National Service imposes an opportunity cost on conscripts. This could be the foregone income they would otherwise have earned, or, more generally, the foregone value of the activity they would otherwise have engaged in. In the case of the (West) German National Service (which has since been abolished – more on this later), estimates of this foregone income range from €2.2bn to €6.7bn.

But these are just static costs. National Service also disrupts young people’s careers at a crucial, formative stage of their lives, which can have longer-term scarring effects. A study of the Dutch National Service (also now abolished) finds losses in lifetime earnings of up to 5% relative to non-conscripts. Similar studies from the US find even larger effects.

OECD-wide, the IZA study finds that National Service reduces economic growth by more than a quarter of a percentage point per year. Given how sclerotic the British economy has been since 2008, an effect of even half that size would be devastating.

This is before we even discuss the productivity losses for the military, and the civilian institutions that employ conscripts. They have a higher staff turnover, and they have higher supervision costs, because those people do not want to be there.

Are all of those countries just doing National Service in the wrong way? Has real National Service never been tried? Could it not be organised in such a way that people pick up valuable skills?

Hardly. National Service is built on an egalitarian ethos. The idea is that every role (or almost every role) within the scheme can be filled by every (or almost every) conscript. This severely limits the scope for skills-based differentiation. We could imagine a two-tier, three-tier or four-tier National Service where some conscripts occupy more skilled roles than others, but that would undermine the whole point of it.

At this point, I can already see the comments from our communitarian friends: “There’s that typical hyper-liberal individualistic mindset again! You guys really don’t get it, do you? You see people as nothing but economic units, whose sole purpose in life is to contribute to GDP. So what if National Service knocks a few pounds off our GDP?? The point is to strengthen community cohesion, and create a spirit of national solidarity! We might be slightly poorer in material terms, but richer in terms of our social fabric. I’m not expecting you to understand that, but some people see our country as more than a big stock market!”

I hate “lived experience”-type arguments, but I’m afraid I have to make an exception here. Because I did National Service, and I can assure you that it does not live up to the romantic communitarian fantasy.

West Germany introduced mandatory military service (Wehrdienst) in the mid-1950s, in response to the threat of the Cold War. This was soon complemented by an equivalent non-military, civilian service (Zivildienst), which was typically done in hospitals or care homes. (Although you could find easier alternatives: I did mine in a youth hostel, which, admittedly, does not sound like a terrible punishment, and I could not claim that it was.)

German Reunification made the original purpose of National Service obsolete, but bureaucracies are slow to adapt, and so, it took another two decades for it to be abolished. Since I finished school in 2000, I still had to do it.

The thing about National Service is that it’s only “national” in a purely administrative sense. When you actually do it, it does not feel very “national” at all. You work at a specific, identifiable workplace with specific, identifiable individuals. To the extent that it creates a sense of community, it does so only among those small groups of individuals.

There were five other chaps doing National Service at the same place with me, and in our case, there was indeed a sense of “we’re all in this together”. It did indeed create a spirit of camaraderie.

However: that spirit did not extend towards wider society – and it definitely did not extend towards “the nation” as a whole. Why on earth should it have? It was based on a shared experience, so by its very nature, it was not scalable.

National Service is a terrible idea from an economic perspective, and a worse idea from a liberal perspective. But even on purely communitarian grounds – it’s not all it’s cracked up to be.