David R. Henderson  

Eliminating Conscription in Singapore

PRINT
Voting, Public Goods, and Free... Single Parenthood: The Reason ...

In response to co-blogger Bryan Caplan, John Smith argues that Singapore needs conscription. Here's where economics is really helpful: in helping us understand the distinction between how big a military to have and how to man a military.

Smith argues that Singapore needs a large military and argues implicitly that to have a large military, Singapore needs a draft. But those are two separate issues. As economists have pointed out, you don't reduce the cost of military manpower with a draft. All you do is keep the government's budget outlay lower than otherwise and, in the process, shift the cost, hide the cost, and raise the cost.

The government shifts the cost onto the backs of draftees, hides the cost by having it not be visible in any government budget but, instead, having it borne by the draftees, and raises the cost by causing manpower to be priced artificially low so that it is used less efficiently.

If the Singapore government wants a large military, it can get it, under a volunteer system, by paying enough to get its desired number of volunteers.

Still, some would worry that the amount required to pay these volunteers would be a budget buster.

Au contraire. Singapore's active-duty military is now about 72,000, of whom about half are conscripts. The conscripts are paid about $500 Singapore dollars a month and up. One Singapore dollar = about 80 U.S. cents. So that's about US$5,000 a year. Of course, just as in the U.S. case when we had a draft, the draft causes there to be less pressure to pay volunteers also. So let's say that the Singapore government ended the draft and doing so caused it to pay an average of an extra $20K per person. I'm betting this is an overestimate. Total expenditure increase per year: 72,000 * $20K = $1.44 billion. Of course it would cost more to maintain a large reserve force also. So let's kick in another $1 billion. Total: $2.5 billion. How does this compare to Singapore's GDP? It's about $240 billion. So the added government outlay to avoid conscription would be about another 1% of GDP. Steep, yes. But not outlandish.


Comments and Sharing





COMMENTS (29 to date)
Steve Sailer writes:

Singapore's ruling class is pretty good with numbers. So, if they are dead set on conscription, it is probably for political reasons. Perhaps, in a multi-ethnic society, the draft allows the government to round up impressionable young men and inculcate Singaporean nationalism in them?

david writes:

How much would it cost to pay 1.2 million healthy males to remain ready to mobilize at a day's notice for two decades?

david writes:

To clarify, I think you can sensibly set forth a case that Singapore should militarize less. I don't think it is sensible to claim that it can meet its current level of militarization through wages, though, since it is spectacularly labour-intensive and Singapore is spectacularly rich. Baumol is not good for the budget.

It's possible you are misinterpreting the number of conscripts at any one time, since the conscripts regularly rotate in and out of the army there. The actual number of people obliged to serve army duty is much, much higher.

philemon writes:

The 72,000 figure refers only to those serving their 2 years--most of which is training. The actual military strength is in the order of 350,000. And like the Israelis, if it really comes to it (and I pray that we never have to), many many more people than that 72,000 will be called up--for active, front line duty, mind you.

Having said all that, I do agree with David that Singapore pays the costs one way or another--either in the conscripts' time and lost productivity, or in an explicit government budget item.

Incidentally, I see a long term shift away from having lots of bodies towards increasing use of technological 'force multipliers'. As it is, the length of the active call up has been shortened from 2.5 to

philemon writes:

(sorry; con't)

...to 2 years. But I don't quite see the SAF moving away from at least three (combines arms) divisions as the backbone of the army.

Essen writes:

Steve Sailer has a point.

Time we allwork in the trenches and not limit the expression to a management jargon.

When war comes to the shores not all laws of economics can stop conscription. Be ready to reinvent the phrase Last man standing.

http://www.beedictionary.com/blog/2011/01/meaning-and-usages-of-conscription/

Motoko writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

John Smith writes:

To David Henderson:

While I am delighted that my comment has been replied to, I am a bit disappointed at the quality of the reply. This is somewhat my fault. I being a student of economics (mere economics minor, though at the flagship national university), thought that Caplan wouldn’t reply and hence I didn’t invest the effort to state my thoughts well. However, I do place the poor quality of the numbers you ran entirely on you though.

I understand entirely the concept of hidden cost and opportunity cost. My question is this.

*** Does the national government enjoy certain inherent abilities by virtue of being a government that allows it to enact conscription at a reduced opportunity cost? ***

Accordingly to Wikipedia (http://en.wikipedia.org/wiki/Singapore_Armed_Forces), the Singapore military has a strength of over 400 000 soldiers, of which over 300 000 are reserve troops. Presumably in any conflict, the garrison troops sent to hold down the enemy cities after they have first been pacified would be predominately reserve troops, which are of lower quality than front-line professional soldiers. The reserve troops are *THE* primary component of the army. They certainly cannot be waved away with a mere billion as you did with no justification for your numbers whatsoever.

Let we assume that the strength of the military is an issue best left to the judgement of the military and the politicians overseeing them. For Singapore to abolish the draft and yet still maintain a military of similar size (garrison troops are heavily dependent on quantity, in particular since Singapore would not be greatly adverse to terror tactics to make up for the lack in quality), Singapore would need to either retain the same military structure of a small core of full-time soldiers and a huge reserve force of reserve troops who are now willingly serving for pay or to have a huge professional army on the order of hundreds of thousands of troops.

With respect to the latter, let me run the numbers for you, but done better than your mediocre effort (If you sense that I am still disappointed, you are right. Numbers are the bread and butter of an economist. Attention to details is important). All figures are in local currency, Singapore Dollars SGD. Assume a professional military of an additional 400 000, excluding the existing 30 000 professional soldiers for the sake of convenience and keeping in mind that there are now no conscripts being paid either.

Assume that the average annual pay is that of an entry level fresh university graduate; the rationale being that the officers would earn more and the grunts would earn less and it would all average out. Do keep in mind that the occupation of being a professional soldier is not looked well upon within Singapore society, unlike in the States, and your suggested average annual pay of 26 000 (effective monthly cash salary of 1700) for grunts and officers as a whole is ludicrous. The annual pay would then be (3000 entry level monthly pay X 13 months of pay X an additional 16 percent of social security payments by the employer known as CPF) 45420. Let us call it 50 000 for easy accounting.

The total annual compensation would then be 400 000 X 50 000 for a grand total of 20 billion. The 2011 government budget is 48 billion (http://www.mof.gov.sg/budget_2012/revenue_expenditure/index.html) with a 2011 military budget of 12 billion (http://en.wikipedia.org/wiki/Singapore_Armed_Forces). The new military budget would then be 32 billion, out of a 2011 GDP of 327 billion (http://www.singstat.gov.sg/stats/keyind.html#econind). In other words, the military budget alone would be 9.8% of the GDP, up from its current 2.7%. The entire government budget would be 20.8%, up from its 14.7%. This would be extremely expensive, not to mention the destructive effects on tax revenues by withdrawing 400 000 males of working age from the economy.

And yet, opportunity cost states that Singapore is already doing it (through conscription instead, not the more expensive and more capable huge professional military outlined above), albeit with the costs hidden. Which brings us back to my initial question. Does the national government enjoy certain inherent abilities by virtue of being a government that allows it to enact conscription at a reduced opportunity cost?

Let us now look at the other alternative I had mentioned earlier of retaining the same military structure of a small core of full-time professional soldiers and a huge reserve force (let us call it 400 000 for simplicity sake) of troops who are now willingly serving for pay.

Military service within this context can be divided into two primary components. The first being the actual state of employment within the military, and the second being liable for mobilisation in the unlikely event of war.

The first would require routine compensation as with any other occupation. The second, however, presents a trickier issue. One can clearly not be called up for service without having first been trained to serve in the military. Untrained and unqualified troops would be nothing more than mere cannon folder, which would be of minimal military value within the context of modern warfare and be politically untenable.

The issue then arises of how to compensate for the liability for mobilisation. You cannot select such individuals only at the point of mobilisation as there would not be enough time for military training. Instead, you must enlist these individuals into the military for short duration training to bring them up to acceptable standards before releasing them from service (presumably with periodic refresher training), with the understanding that the compensation includes mobilisation liability.

Keeping in mind the lack of patriotism in Singapore, this would effectively be an option on your probabilistic life (since serving as garrison troops does not guarantee death, but is only probabilistic in nature), callable in the event of war. Unlike the scenario with a full-time professional military of 400 000, the state is explicitly paying you solely for the right to mobilise you in the event of war, i.e. your probabilistic life.

Now, I am sure that out of a population of 5 million (http://en.wikipedia.org/wiki/Singapore), there would be a few individuals that are willing to take this gamble. Males fit for military service as of 2005 is only 934 000 (http://en.wikipedia.org/wiki/Singapore_Armed_Forces). Assuming half of the population is male and half of the male population is in the work force, this would yield 1.25 males of military age, roughly in line with the smaller figure of 934 000 for the earlier year of 2005. Let’s call it an even million.

Do you think that 400 000 out of this million, or 4 out of every 10 men, would be keen to sell a option on their probabilistic life, callable in the event of war, for any reasonable price affordable by the state? 40% seems rather high, don’t you think? I have made several generous assumptions on my behalf, so let us now revise it downwards to 20%. Does 20% not seem too high as well? Bear in mind that the state cannot afford to pay an astronomical amount such as one million dollars each to convince people.

It seems to me that the state enjoys certain inherent abilities by virtue of being a government that allows it to enact conscription at a reduced opportunity cost. By forcefully seizing the private property of mobilisation liability from Singaporean males without compensation, the state is firstly ensuring that it is able to retain a viable military of sufficient size (which may otherwise be explicitly costly, without conscription hiding the cost, for a huge professional military, or problematic in terms of recruitment for both a huge professional military, and a small professional military with accompanying huge reserve force).

Secondly, as economists, we are well aware that humans are irrational. Could it not be that something that would cost a huge amount if paid explicitly (buying an option on your probabilistic life), would only cost a small amount (if we monetise the psychological distress of the state seizing and holding the option on your probabilistic life) if imposed implicitly (through conscription)?

After all, out of sight, out of mind. It is one thing to join the entire cohort of males of your birth year in compulsory military enlistment, or even to sign up willingly as a professional soldier (while quietly lying to yourself that the government would never have to call up its option on your probabilistic life). It is quite another thing to so cold-bloodedly sign over your probabilistic life so explicitly and visually.

I would greatly appreciate your insights on this, and very much that of Caplan.

*** Does the national government enjoy certain inherent abilities by virtue of being a government that allows it to enact conscription at a reduced opportunity cost? ***

My apologies in advance for any errors in my numbers or logic.

Ken B writes:

I intend to use this to argue against Obamacare.

I often make a simple argument to people: when we have to allocate limited resources markets are the best social technology we have. "What part of optimal do you not understand?" is how I sometimes provocatively put it. It's nice to have examples. This is a terrific one. Most people I debate with will oppose the draft, and see it (rightly) as oppressive. This kind of careful analysis will help them see exactly why and how, because it's quite analogous really in spreading costs, and hence more see clearly why and how markets are good for such questions.

Peter H writes:

John/Bryan

It seems the main issue is the size of the reserve forces as opposed to the active duty forces. My question would be to what extent reserve forces could be replaced or supplemented by mercenaries? Singapore is a wealthy country, and in the event of a war which threatened its continued existence, would presumably be willing to deploy as much of that wealth as needed to stay around. I don't know how large the mercenary industry is worldwide, but I know it's non-zero. Would it be possible to buy options/retainers on hiring mercenaries like Blackwater (or whatever they changed their name to)? If so, could it sufficiently provide Singapore's security needs, or does a foreign force have inherent problems. The issue you present most compellingly is that you need an option on an insane portion of the Singaporean population's lives to have a large reserve, but if you expand the world of possible reservists to the populations of a few large countries that have good relations with Singapore (I'm not geopolitically savvy enough to say whom, but I doubt you'd want to hire Indonesian mercenaries), it seems to get around that reasonably elegantly.

Peter H writes:

Somehow I mus-addressed that to Bryan instead of David, apologies.

David R. Henderson writes:

@John Smith,
My apologies in advance for any errors in my numbers or logic.
No need to apologize. If you have read this blog much, you'll know that I have often made mistakes, not usually in logic, but occasionally in numbers. :-)
Now to the specifics. Two problems with your numbers:
1. I did NOT assume that officers would be paid only $26K. You and I both understand that this is ludicrous. I suggested that everyone would get a $20K increase, which is a very different proposition.
2. I think it's unreasonable to assume that you need to pay reserves the same full-time pay you would pay people on active duty. I think a better estimate would be about one tenth of full time pay. So if we take your $50K, which is about U.S.$40K, times one tenth, we get $4K annually for reserve pay. The number of reserves is about 350,000 (see my link in my post). So we get $1.44 billion in annual reserve pay. So my gut feel number of $1 billion was pretty close and WAY closer than your estimate.

David R. Henderson writes:

@Ken B,
Thanks.
There's an even better argument that economists made about the inefficiency of the draft: it put, in words that were the title of an anti-draft book by Bruce Chapman, "The Wrong Man in Uniform." That is, it had no good way to distinguish between a high-supply-price person (Elvis Presley was the extreme example) and a low-supply-price person. I didn't use that argument in this case because with a close-to-universal draft, that argument loses most of its force.

david writes:

I dispute that $4000 annually would be enough to convince Singaporeans - a rather wealthy lot - to agree to be called up every year. This is close to one month's average income, but the period for which they are called up is already three-ish weeks, if I am googling their obligations right.

Regardless - surely it is not fair to compare the deadweight costs of conscription to an assumption of no deadweight costs in raising taxes?

Chris H writes:

Hey John Smith!

Sorry I'm not Dr. Caplan or Dr. Henderson, but I do think I have some things to note about this argument. Let's start with this:

It is quite another thing to so cold-bloodedly sign over your probabilistic life so explicitly and visually.

This to me is one of the major lynch pins of your argument. Wouldn't the price needed to convince people to give the Singaporean state the right to call them up to risk their lives be astronomical? I think that very much depends on the level of risk involved. People can make irrational judgments, but those tend to be rare as the consequences for being wrong tend to be serious (there are notable exceptions but I don't think this is one given that being wrong on overestimating probabilities means lost income and underestimating means having to fight in a war, now the consequences are asymmetrical but what I'm going to argue going forward is that the chances of over vs underestimating are even more so, the chances of being called up in Singapore's case are so low that even massive overestimations of risk would still lead a person to conclude the chances of being called up are low and thus not demand absurd amounts of compensation). Now values such as patriotism can play a role but it doesn't seem plausible to me that the cost curve would be bent so far downward as to make such a system completely unaffordable.

For one thing, it's worth remembering patriotism is not sufficient to eliminate the idea that a person is gambling with his life. If patriotism blinded people to that fact then you would not expect there to be significant drops in recruitment based on whether the country was at war. Or if individuals found it more daunting you wouldn't expect non-parental influencers to drop in support of people to join the military (after all they aren't risking their own lives or those of their children) and yet even they show decreases in recommending military service (http://www.military.com/NewContent/0,13190,Philpott_080405,00.html). So patriotism does fail to blind people to probabilities, but how big is this effect? Perhaps we can find an example of people risking their lives with no self-deceiving effects of patriotism. Luckily we have just such examples.

Consider this: whenever a company pays you to drive somewhere they are taking a probabilistic option on your life with absolutely no veneer of patriotism. Remember that car accidents are the 6th highest cause of death in developed countries (http://en.wikipedia.org/wiki/List_of_causes_of_death_by_rate#Developed_vs._developing_economies). The chances of dying in such an accident while on the job are thus very real and people know this. So how much do drivers have to be compensated to induce them to take the very real chance of dying simply for their job? About 55 cents a mile (http://www.bizjournals.com/wichita/blog/2011/12/irs-to-keep-standard-mileage.html).

Now, you may object that while yes the chance of dying in a car accident is real and widely known, that chance is also partially in the driver's control and is low. To the first objection, it need only be noted that the exact same is true for war. You are far more likely to survive a war by behaving more cautiously than by being a "hero" and charging enemy lines. Similarly you're far more likely to survive driving by driving defensively than by getting drunk and swerving all over the road. And in both cases, there is always the possibility that your actions may be as safe as possible and you still get unlucky and die. What an all volunteer force also does however is create an incentive for people to sort themselves out on their ability to survive a war. Those who are better with guns, have firmer grasps of military tactics, or are simply more callous and calculating are more likely to decide signing up for reserves is worthwhile. Similarly, people who trust their driving skills more are more likely to sign up for jobs which require a lot of driving. This partial control of risk therefore tends to enhance quality over draft situations.

To the second objection, we need to consider two factors: what are the chances Singapore will go to war and what are the chances a soldier in that war will die. Beginning with the first, three factors to consider are Singapore's history of going to war, the likelihood of anyone invading Singapore, and the likelihood of Singapore invading or participating in the invasion of another country.

For the first there would be lots of reason for optimism. Singapore has not been significantly involved (certainly not to a level requiring calling up reserves) in a war since independence from Malaysia. It did commit a small numbers od troops to the wars in Iraq (http://en.wikipedia.org/wiki/Multi-National_Force_%E2%80%93_Iraq#List_of_countries_in_the_coalition) and Afghanistan (http://en.wikipedia.org/wiki/International_Security_Assistance_Force#Non-NATO_and_non-EAPC_nations), but both times with in capacities which resulted in zero combat casualties (For Iraq: http://en.wikipedia.org/wiki/Casualties_of_the_Iraq_War#Coalition_deaths_by_hostile_fire and for Afghanistan: http://en.wikipedia.org/wiki/Coalition_casualties_in_Afghanistan). So not only does Singapore have a fairly peaceful history, the only cases it has actually participated at all in military actions 1) never even came close to requiring calling up reserves and 2) never led to a single combat casualty. But the future may be different so what do the prospects for future wars look like?

Considering the possibility of invasion, the prospects again look optimistic. Singapore has minor territorial disputes with Malaysia and Indonesia, but there is no indication that any side in those disputes is seriously considering invasion to resolve them. Beyond that there is little reason any country would want to invade Singapore. Singapore maintains a unilateral free trade regime meaning there is no advantage to be gained from more open markets from invasion to any country. Singapore doesn't sit on any major natural resources meaning there's no advantage to be gained from seizing monopolistic control over a resource. Attempting to seize Singapore for control of the Straits of Malacca is also unlikely given any country which sought to do that would face major opposition from major powers like China, Japan, or the United States. Invasion of the city-state is thus highly unlikely.

What are the chances Singapore will invade another country? The participation in Iraq and Afghanistan show that Singapore might participate in such an invasion, but only to a minor degree with no chance of reserves actually being called up. Such wars need not worry any person considering signing up for the reserves. A more major invasion on Singapore's part also seems unlikely. Singapore has no strategic depth meaning an invasion of any country with any ability to strike back would be disastrous as said country would bring fighting directly to the city of Singapore itself. Invading a country with no such ability still makes little sense from the leadership's perspective given that their free trade regime gives them access to any imports they might require and expansion of Singapore's territory might mean making a future invasion more likely. This is not even to mention that Singapore's elites are intelligent and understand just how serious the costs of war would be against any benefits they might receive. The chances of an offensive war requiring reserves is thus also extremely low.

Now what about the chances of dying should a war happen? This very much depends on the nature of the war, but modern wars tend to have fairly low deaths-to-wounded ratios due to modern medical technology. In the war in Afghanistan and Iraq have resulted in US death-to-wounded ratios of about 1 in 10 (http://en.wikipedia.org/wiki/United_States_Forces_casualties_in_the_war_in_Afghanistan and http://en.wikipedia.org/wiki/Casualties_of_the_Iraq_War#U.S._armed_forces). Overall the chances of dying as a member of the American military who has served in Iraq, Afghanistan, or both is about .3% (taking the number of dead divided by the number who have served in those wars which can be found here: http://abcnews.go.com/Politics/us-veterans-numbers/story?id=14928136#1). Death rates for the conventional portions of these wars is even lower (at .1% for US troops: http://en.wikipedia.org/wiki/2003_invasion_of_Iraq). These were wars against a belligerents with relatively weak militaries however and death rates would likely be higher among more evenly matched opponents. One example that might be more interesting for determining casualty rates between relatively evenly matched opponents would be the Libyan Civil War (http://en.wikipedia.org/wiki/Libyan_civil_war). Rebel rates depend upon how reliable the NTC's estimates for the number of volunteers by the war's end is so instead I'll focus on the less variable Libyan military's numbers. Gaddafi's forces by that measure had between a 10-20% death rate. That's definitely significant! But once multiplied by a near-zero chance of being called up to fight, most people won't even bother getting to the point of considering their chances of dying in a war. If the chance of a war is almost zero the chance that large numbers of people see this as nearly free money is significant. Furthermore, because Singapore is a democracy, those signing up would be able to anticipate a policy shift towards more belligerence and adjust whether they'll sign up for reserves accordingly (which creates the added benefit of decreasing the probability of unnecessary bellicosity).

Now, Singapore has a very economically literate elite class. They almost assuredly can figure all of this out. So why maintain conscription? I think Sailer might have a point on their motivations. Military service as a means of enhancing loyalty to the state through inculcation of the values of the Singaporean state makes sense. What may be necessary then to convince Singapore to drop conscription is arguing that such value uniformity is unnecessary and less valuable than the gains to be had from switching to an all-volunteer force.

Ken B writes:

DRH:

There's an even better argument that economists made about the inefficiency of the draft: it put, in words that were the title of an anti-draft book by Bruce Chapman, "The Wrong Man in Uniform."

Great point. But ironically that is almost always presented as an argument FOR the draft. "We must all do our bit". I hear more umbrage about being able to buy sustitutes in the civil war (or $300) than I do about the draft itself. A misguided sense of egalitarianism; misguided because it fails to allow people to set their own priorities for risk and reward.
Another nice example as you note of the ill effects of cutting out prices and bargaining.

cyl2002 writes:

I think many of the discussions miss the point for conscription in Singapore. Conscription is not going anywhere any time soon. Economy or inculcating value are certainly factors but there are other compelling reasons as well.

One important consideration is the deterrence factor. Any potential aggressor will soon realize that it has to content with a country where every able-bodied man is trained in the military. Almost every male is able to pick up a M16 rifle and use it.

John Smith writes:

To Peter H:

The answer is to a negligible extent. The geographic nature of Singapore’s security position is a very poor one. The country is extremely small and a defence in depth is impossible. Even light artillery can readily fire into the government centre and most population centres from within the borders of our enemies.

Imagine your country is New York City and New York State is your enemy. Our situation is significantly worse than that. There are only a few narrow approaches to and from our country. In the event of war, the overwhelming objective is for our military to secure these approaches and break out from there to secure a buffer area as a staging area, from which to advance upon the enemy.

We certainly cannot wait for Blackwater to respond to our callup, not to mention that private military companies are extremely untrustworthy. Blackwater is an excellent example.

John Smith writes:

To Chris H:

I think people are more irrational than you give them discredit for.

The company is paying you for driving a vehicle. Let’s say the company had presented it to you as being that the self-driving vehicle would drive itself with the average skill level of human drivers and that in the event of an accident that would have been otherwise fatal for human drivers, you will be executed. Would you still take up the job when it is presented that way even though it has exactly the same effect on your life?

Do bear in mind that I by no means discredit your ideas. But the proportion of people that we would require to sign up, 4 out of every 10 men, is so high that I find it difficult to believe we can achieve this.

John Smith writes:

To David Henderson:

1) A 20k increase to the existing salary of 6k (the conscript officers are only paid slightly better than the grunts, no significant difference) is an overall salary of 26k. The conscripts are led by conscript NCOs and conscript officers. So, your average salary of 26k would have to be shared out among NCOs and officers and grunts, which is ludicrous. It is not 26k for officers alone; it is an average for grunts and officers as a whole.

2) I did not indicate that the reserves would draw full time pay. The estimate of 32 billion is for a huge full-time professional military (which would bring Singapore into the region of being close to a weak Great Power, Germany, instead of the Middle Power it is now), which I took as one of two possible options. You misunderstood this.

My objection to the other option of a large reserve force is that people would not willingly sign up for any reasonable amount of money in the numbers that would be needed. You own country’s recent experience should tell you that invading another country has a huge manpower draw. And in any case, we had agreed to defer to the military on judgement of military affairs.

You are assuming that 4k is enough to meet the reserve price of every single one (or at least the overwhelming majority) of the conscripts. The entire point of this discussion about the draft is that different individuals have different price points. What on earth makes you think that 4k is enough to buy off everyone? Keep in mind that by my rough calculations, 4 out of every 10 men who are even remotely fit for military service would have to willingly sign up as reserves. 4k USD is around 1.67 times of the average monthly wage (2011 average wage for the population was around 3k SGD).

This is just as ludicrous as your notion of 26k average pay. 4k USD is not even a month’s wage for me. Let me assure you that under no circumstances would I willing place myself under military law and discipline (You cannot very well conduct refresher military training under civilian law. You might as well save the trouble and not train), refresher training (even if the refresher training was paid in addition) and the prospect of losing my life. Would you, a high ranking economics professor, be eager to lay down your probabilistic life for 4k USD?

It is really this exact issue that is troubling me. Caplan and you keep insisting that conscription can be readily abolished. And I, the good economics student, really and truly want to believe you. But exactly how would it work in the real world? Specifically how? Not in broad vague general terms. But in concrete specific terms with real exact numbers. You guys are the numbers men, aren’t you? Give it to me straight.

Chris H writes:

John Smith writes:

The company is paying you for driving a vehicle. Let’s say the company had presented it to you as being that the self-driving vehicle would drive itself with the average skill level of human drivers and that in the event of an accident that would have been otherwise fatal for human drivers, you will be executed. Would you still take up the job when it is presented that way even though it has exactly the same effect on your life?

I think there are two reasons why someone would rationally refuse to take that job. First, anyone believing to be above-average in terms of driving or capable of improving their skills to being above average would be needlessly increasing their chance of dying by accepting this driverless car deal. Remember that a part of this argument is that we would expect people with more driving skills to be more likely to take these jobs. In order to induce a person to accept this you would have to convince them that the car has at least their skill level in driving. Given that the penalty for being wrong in one's self assessment of this car's ability is potentially very high, people would be rational to be skeptical of such claims without strong, trustworthy evidence. Think of this like chaffeurs. People have the chance to observe a chaffeur's driving and reckless chaffeurs are not hired. I could imagine a similar policy occurring here. Assuming that you could manage to produce such evidence however there is a second problem: the signal that such a policy sends about an employer.

A policy of killing an employee when the self-driving car they are associated with enters in a fatal crash is non-rational. The company gains nothing by adding an employee death to the car's damage and therefore any potential employee upon hearing of this policy would justifiably assume that the company was run by insane people. People don't like working for insane bosses. When a boss asks an employee to risk his/her life for the job (even if it's only a small risk) that employee accepts said risk only if undertaking that risk is an aid to the job that is worth it. An employee will accept risks which actually can be shown to be necessary to improve the job. But an employer who puts on risks to employees that do not improve the job is going to run that company into the ground financially paying unnecessary risk premiums and losing employees. It would be rational therefore for an employee to reject such an employer even with higher risk premiums.

Thus in order for this counter-example to work, we have to assume a proven driving level at least at the employee's level and then assume that there is a necessary reason why the employee would have to die in the event of a fatal car crash. In such a scenario here's what I would predict. In the early stages there would be higher risk premiums needed to induce people to use this technology than regular driving. There is a rational reason for this as new ideas are more likely to have problems that need to be worked out and therefore fewer people will take a chance on newer ideas than tried and tested ones. But once the system is well-established and tested, I would expect people to view it as not substantially different from driving, only with the added benefit of not actually having to leave the office. Overall there is a tendency towards rationality in people as shown by the everyday smooth functioning of, well, everything. If irrationality was the norm, then we would expect that most car drives would end in crashes, most prices for everyday goods would fluctuate hugely and widely, and most people would engage in non-pleasurable, non-renumerative risky behaviors like seeing how much lead they could intake before dying.

A mostly irrational society wouldn't be just a little bad, but would self-destruct almost instantly. That is why there is an assumption that people tend to be rational and any kind of persistent irrationality requires an explanation.

David R. Henderson writes:

@John Smith,
I think we're at the point where the main people paying attention are you, Chris H, and I. So I don't want to take time to reply in the comments.
Since this issue has generated such discussion, though, I'm thinking of writing a follow-on post. Just to make sure I know what we're arguing about, are you arguing for a Singapore draft? Also, do you still live in Singapore?

Ken B writes:

I hear a lot of people sing the praises of compulsory service after high school, along the lines of many EU countries. Not necessarily military, perhaps road work etc. I am always surprised by how many people think this a great idea! If you do a follow up David, please touch on this issue.

Matt G writes:
Wouldn't the price needed to convince people to give the Singaporean state the right to call them up to risk their lives be astronomical?
I don't see why - people regularly join military organizations, giving them the right to require them to risk their lives and giving up many of their freedoms, for monetary compensations that are not at all astronomical.

The more subjective reasons why people do so are harder to evaluate.

I suspect that from the perspective of Singapore's political elite, as Steve Sailer suggested, the primary benefits of conscription are political and psychological. If you already accept that authority has the right to your time and life, why quibble about bubble gum disposal fines?

John Smith writes:

To Chris H:

My counter example has those assumptions you indicated, in that there is a rational basis for executing the drivers.

I agree with you that the overwhelming amount of time people are rational. However, in the case for Singapore, when it is portrayed in such stark terms, I think that people may demand somewhat of a premium over what is rational.

More importantly, as I mentioned, the sign up rate would have to be 4 out of every 10 men who is even remotely fit for military service. Do you think we can meet the threshold? Without bankrupting the state.

John Smith writes:

To Matt G:

If you were addressing me, that is not an accurate description of my position, nor did I made that statement.

John Smith writes:

To David Henderson:

Excellent. Please do so. It is so annoying when Caplan persistently and repeatedly insists that we should abolish conscription in Singapore. But not once (to my recollection) has he actually described in detail the mechanism for doing so.

To clarify, I am not necessarily arguing for conscription. I see myself as being one of the intellectual elite and I believe firmly in the tenets of economics teaching. As such, I feel as though I should be convinced that it can work out. I just don’t understand how. This is where you, an economics professor, come in. I am implicitly assuming that the draft may possibly be more efficient because of human irrationality on the part of the population towards so explicitly selling their probabilistic lives and difficulties in recruiting hundreds of thousands of reserve troops under these conditions.

Yes. I am born, raised and living in Singapore, as are my parents. Not a drop of non-Chinese blood in me going back for centuries either (as far as I know). I am as pure a Singaporean as you can reasonably find. My forefathers didn’t come over on the equivalent of the Mayflower though.

My request is this. To specifically and clearly describe the exact manner in which Singapore may abolish the draft (and yet retain a massive military force of hundreds of thousands available for callup) and explain this mechanism in great explicit details, including running all the numbers for doing so and providing credible sources and basis for said numbers.

All assumptions and logical basis must be reasonable within the context of Singapore. No handwaving and insisting on the existence of patriotism or what have you. Note that military training of around a year (essentially abolishing the second year of unit deployment and only retaining the current first year of basic and intermediate training) would be the bare minimum needed to bring the raw recruits up to standard before they can be discharged into the reserve force, and that periodical refresher training is required.

[comment edited with permission--Econlib Ed.]

Chris H writes:

John Smith writes:

I agree with you that the overwhelming amount of time people are rational. However, in the case for Singapore, when it is portrayed in such stark terms, I think that people may demand somewhat of a premium over what is rational.

More importantly, as I mentioned, the sign up rate would have to be 4 out of every 10 men who is even remotely fit for military service. Do you think we can meet the threshold? Without bankrupting the state.

If people do demand somewhat of a premium then here's what I would expect. If the premium is relatively minor there's no problem. If it's larger then this creates incentive for advertising/propaganda campaigns to obscure the issue and bring down premiums to a reasonable level. This has worked for creating patriotism for many countries in the past so I don't see why it couldn't work in Singapore.

Could you get 40% of the male populace to sign up affordably? I'd say probably as long as Singapore remains a non-belligerent country with few threats. In such a case this is nearly free money for the individuals. In the event that threats begin rising then the conscription debate can be re-opened.

John Smith writes:

To Chris H:

The reason is because there is no patriotism in Singapore. Singapore society is nearly an ideal society by economic modelling standards. We are a nation of cold emotionless robots, that solves our individual utility functions to achieve the highest utility curve in anything we do.

And yes, I get that this seems contrary to my earlier position. If we are so rational, why will we demand such a premium? Can’t the government just use figures and facts to convince people that the premium is undeserved after all?

Well, exactly how would this work? You are going to convince people that it is totally cool. We are not going to screw you, most likely it is safe? Leaving the public’s distrust of the government aside, isn’t this just going to bring home all the more the harsh and “cruel” nature of this exchange? By trying to convince them it is not “cruel”, you are convincing them even more that it is in fact “cruel”.

It is not free. You still need to attend annual refresher training. You can’t expect people to remember how to use a rifle after a few years, much less small unit tactics. I can assure you that it has been so long since my last call-up that I can’t remember a damn thing.

Comments for this entry have been closed
Return to top