Tuesday, May 24, 2011

Seeing the Poor as Investors

Sir Partha Dasgupta
Upon reading Sir Partha Dasgupta’s extremely informative book on development economics, An Inquiry into Well-Being and Destitution (Oxford: Clarendon Press, 1993), I came across the following very interesting insight (pp. 245-248).

Development economists, NGO’s, and assorted concerned policymakers often say that one of the pressing obstacles to bettering the condition of the very poor in developing countries is a lack of capital accumulation that can lift them out of the dreary cycle of living hand-to-mouth. If you cannot save any money because all of your income goes to immediate consumption — to keeping yourself fed and clothed — then you cannot save anything that can be invested in a way that will help you to move up the economic ladder. Low wages, at least at the extreme of poverty considered by Dasgupta, are a trap.

Dasgupta doesn’t argue with the sad fact that extreme poverty is a trap, but he offers a different way of conceptualizing it. Standard economic models of savings and investment, he says, don’t translate well to poor peasant economies, where “savings” are not necessarily composed of money kept in a bank, and where much “investment” doesn’t pass through one either. When a peasant spends his meagre wealth (or his labour, which is much the same thing) to terrace or irrigate a field, the standard models view this as consumption, whereas the peasant obviously views it as investment. Looked at from the peasant-level point of view, there is a lot more investment going on in the peasant community than is captured using the standard economic models, whose measurements rely on savings and investment tied to financial and other economic institutions.

As poor as such peasants may be, there are those who are still poorer. Consider, for example, a day labourer in India, who owns no land, and whose wages are just enough to support his continued existence and no more (or if you prefer Marxian terminology, his wages are only enough to reproduce his labour). Again, for this day labourer, the conceptual line between consumption and investment is very blurry. When he spends his wages on rice to keep body (but not soul) together, the standard economic models treat this as consumption. But for the labourer, it is an investment in the only capital he owns — his labour. The simple fact is, if he doesn’t eat, or if he doesn’t eat enough, he will not be able to earn.

(Elsewhere in his book, Dasgupta notes a related phenomena about the extremely poor: their productivity is particularly path-dependent on their nutritional history: a recent history of nutritional deficiency — which could be caused by a short period of lost wages due to a work slowdown — will represent a lost ability to work, which may result in more lost wages and further lost nutrition. But that is just another of the many tales Dasgupta can tell you about the lives of India’s poor.)

Because of their failure to see that there is a blurry line for the very poor between consumption and investment, standard economic models characterize the plight of these people as an inability to save enough money to escape poverty. In short, the main problem of the poor is presumed to be insufficient capital, because what they get all goes to consumption. According to these models, the solution is for countries to increase their savings rate. This is a macroeconomic solution that according to Dasgupta, misses the point. Increasing the savings rate is fine, except that the only people in the country who are likely to increase their savings rate are those whose incomes are already high enough that they can plausibly save more. The very poor don’t have that luxury.

Dasgupta’s alternative way of seeing the situation of the very poor is that they actually do invest rather than just consume, although this isn’t captured by the standard models. As a matter of fact, viewed correctly, the poor spend a greater proportion of their income on investment than the more affluent do. The poor do not suffer from a low savings/investment rate; as a proportion of income their rate is actually higher than average. Their real problem is that their return on investment is very low. In the case of the day labourer, what this means is that, in the very limiting case, his investment (the bowl of rice he eats) only enables him to earn enough to buy another bowl of rice. In other words, his return on his investment is effectively zero. He is doing no more than barely replacing his capital. If his wages were increased, then the return on investment of his bowl of rice would increase too.

It’s not immediately clear to me what policy difference would result from adopting Dasgupta’s view. However, I think there are a couple of possibilities. First, at the very least, his model would have us stop believing that the very poor spend everything they have on consumption. This might in itself be a good thing, since such a mistaken belief can too easily veer into the more ignorant assumption that the very poor are improvident wastrels, incapable of industry or foresight. In truth, they are just as entrepreneurial and industrious as anyone — technically even more so.

Second, and in a somewhat similar vein, I imagine that seeing the very poor as investors of a sort might make institutions more willing to lend them capital, which the poor can then use to increase their return on investment. This is part of the motivation behind microfinance schemes, which are predicated on the idea that the poor in developing countries are not necessarily bad credit risks. Indeed, given the opportunity, their credit might turn out to be exemplary. Microfinance has by most accounts proven very successful at helping the poor.

Third, the orthodox assumption that the national savings rate needs to be increased might actually be detrimental to the very poor. If the only people who actually can increase their savings rate (i.e. the more affluent) were to do so, the immediate effect might be that many very poor people would lose their livelihoods, which are gained from the consumption and improvidence of the more affluent.

At the very least, Dasgupta’s conceptual shift is worth pondering. The economics taught in Western universities might be more parochial than we care to imagine. Its theories might not translate well to other social milieus and may be better at catering to the needs of the affluent than to those of the poor, even when that is not the intention.

Tuesday, May 17, 2011

Slowing Government’s “Impetuous Vortex”

James Madison, republican theorist
Canada’s dismal federal election is over, and I for one have never been more ashamed to be a Canadian. Thanks to a rotten and archaic first-past-the-post (FPTP) electoral system, Stephen Harper managed to achieve a “majority” government with just under 40% of the popular vote. Stephen Harper, whose government has been the only one in the global history of Westminster-style parliamentary democracy to have been held in contempt of Parliament. I won’t delve into his myriad other failings, which range from the dishonourable to the downright criminal, as it would take a book — or more likely a twenty-volume set of them — to comprehensively lay his misdeeds before the weary reader.

An American friend asked me to explain why I was so appalled at the election outcome. I had to put it in terms of his electoral system, by having him imagine that a proven traitor to his nation had just been elected President of the United States, but without a Senate or House of Representatives to keep him in check. That, in effect is the state of Canada’s political system at the moment.

I can complain until I am blue in the face about the undemocratic nature of Canadian “democracy”. Nothing will change, because FPTP is very hard to get rid of. First, there is the challenge of making any alternative system of voting seem appealing to a tradition-bound population with an aversion to fundamental change. Second, there is the difficulty of actually implementing such change within a constitutional framework that makes amendment nigh impossible. Third, any alternative system offered faces the equally daunting challenge of being simple enough for a plurality of the people to actually understand it, no mean feat in a populace as politically ignorant as your average Canadian. Fourth, any government with the power to seriously propose and implement an alternative system of voting will necessarily have come to power through the FPTP system first, and would therefore have little or no incentive to change a system that has by definition worked splendidly for them, thank you very much. And even if they were elected on a platform of change, once in power they can easily engineer the failure of any promised referendum on the matter.

In my mind, the worst part about our “democratic” system is not its inability to adequately represent the will of the people. After all, in those cases where the will of the people is ignorant or degenerate, non-representation of the people’s will would be a blessing in disguise. Rather, I am bothered by the Canadian system’s lack of republicanism. Although it bothers me immensely that we have an immoral traitor like Stephen Harper occupying the Prime Minister’s office, it certainly wouldn’t be the first time in a democracy that something so base has somehow slithered into the seat of power. What really worries me is that once he is “elected” with a “majority” government, we have no countervailing constitutional means of resisting his infernal machinations. The House of Commons (the body most analogous to the US House of Representatives) is a creature of party: Mr. Harper’s party now absolutely controls it, and Mr. Harper himself absolutely controls his party. And since more and more power over the years has concentrated itself in the cabinet, and in the Prime Minister’s Office in particular, the House is little more than a stage for second rate orators to show off their limited political talents for the consumption of the folks back home (most of whom couldn’t care less). What passes for debate in the House of Commons is full of sound and fury, and truly signifies nothing.

Our Senate shares with its US counterpart nothing but its bare name. A Canadian senator is not elected. A Senate seat is a patronage plum given to superannuated political cronies of the Prime Minster. Lately, Mr. Harper has used it to reward hack journalists who report favourably on government activities (or ensure that said activities disappear down the memory hole, where favourable reporting is simply not possible). In any case, by constitutional convention, the Senate is barred from acting as a real barrier to legislation. It is little more than a rubber stamp. Much talk goes on in Senate committees I’m sure, at least among those few Senators who show up for work, but when the chips are down, the Senate will prove worthless.

Canada does have a Supreme Court with a respectable record of upholding civil liberties, but the Court has never before had to face the sinister brutality of a Prime Minister like Stephen Harper. Alexander Hamilton noted in Federalist No. 78 that the judiciary is the weakest branch of government, having “no influence over either the sword or the purse... and must ultimately depend upon the aid of the executive arm even for the efficacy of its judgments.” Harper has ignored the judgments of the courts before (e.g. the Omar Khadr case), and has railed against the “tyranny” of unelected judges in the past. I don’t see why the leopard should change his spots now. Consider also that there are expected to be several Supreme Court vacancies over the next four years, and these appointments will be in the gift of the Prime Minister. Two Supreme Court justices last week announced their intentions to retire. I expect that their seats will be occupied by new justices with more pliant bums, ones who hold similar views to the Prime Minister’s on the need for unelected judges to keep quiet about things like rights or the rule of law and its supremacy.

Might we be in the unenviable position of having to look to the Governor General to protect our freedoms from the growing despotism? I say unenviable, because the Governor General is an unelected representative of the Queen of Great Britain, a regrettable relic of our days as a British colony. Although our current Governor General is a noted constitutional scholar, there are three reasons not to expect any help from that quarter. First, as with the judiciary, his is a weak branch of government, controlling no resources, and having little democratic legitimacy. If push came to shove, he would get shoved very roughly indeed by a bully like Mr. Harper. Second, he is the representative of a foreign monarch who I have no doubt cares little very little for the fate of our nation, and who is relatively powerless even in her own country. Any real resistance by the Governor General would end up being the fast track to abolition of the monarchy in Canada. Third, again by constitutional convention, the Governor General’s role is mostly symbolic (of what?) — he is a sort of Master of Ceremonies and a rubber stamp for legislation passed by the government.

To reiterate, I’m not so much bothered by the lack of democracy in our system as I am by its lack of republicanism, its lack of counterbalances to executive power. It is possible to have too much democracy, just as it is possible to have too little. Which begs a question: What, precisely, is the difference between a democracy and a republic?

Democracy vs. Republicanism

Much has been written, and much of it opaquely, about what the difference is between a democracy and a republic. I will offer my own cryptic definitions of each, and then expand on them a bit, hopefully making them a little less cryptic.

Democracy is self-determination by the people. This may be done directly, as in a small city-state like ancient Athens, or through representatives, as in most large modern democratic assemblies today.

Republicanism is self-determination by the people as their better selves (in a sense to be explained below).

Democracy has a normative grounding, embodying the concept that the sovereign people are the ultimate source of political legitimacy. Since they are, after all, the ones who will be subject to the coercive powers of government, the people ought to have a say in how those powers will be wielded. On the other hand, republicanism is doubly normative, for it is not merely that the people are the source of political legitimacy; rule by the people is given an added legitimacy by virtue of their being capable of governing themselves. If democracy is good, republican democracy (or a republic with a strong democratic element) is better still.

But what exactly do I mean when I speak of rule by the people as their “better” selves? Do I mean rule by the better sort of people? This has been a common view throughout history and is the concept at the root of aristocracy. This is not what I have in mind. Nobles have proven themselves to be just as fallible, just as selfish, as the common people. Perhaps the only advantage of an aristocracy is that its members have the leisure to stop and consider things. This does not mean that they will stop and consider, or that they will consider well. However, it might partly explain why many republican forms of government have what could be called an “aristocratic” element, a body of people who are not constantly grubbing after votes or currying the favour of a prince. This is usually because either they are not elected, or are elected for long terms, or do not hold their offices at the pleasure of another on whom they are dependent. There is a place for such a body, but its members do not have special or mystical powers such that they should be the sole rulers in a state. If leisure is the main source of whatever political aptitude they bring to the table, then this also implies that the “aristocratic” element does not form a natural ruling class, since leisure is not transmitted through blood or genes.

Madisonian Republicanism

The question is still unanswered: what do I mean by republicanism as self-determination by the people as their better selves? I ask the reader to consider the following brief passage from Tocqueville’s Democracy in America: “What the word ‘republic’ means in the United States is the slow and tranquil action of society on itself. It is an orderly state truly based on the enlightened will of the people. It is a conciliatory government, whose resolutions ripen slowly, are debated deliberately, and are carried out only when mature” (New York: Library of America, 2004, p. 456). There are two key expressions here. The first is “enlightened will of the people”. This could be Tocqueville’s equivalent to my expression “people as their better selves”. But we are still left wondering what is meant by “better” or “enlightened”. The second key expression is more helpful: “debated deliberately, and… carried out only when mature”.

Tocqueville’s view was later echoed by Abraham Lincoln: “A majority held in restraint by constitutional checks and limitations, and always changing easily with deliberate changes of popular opinions and sentiments, is the only true sovereign of a free people.” (First Inaugural Address, March 4, 1861). Again, as in Tocqueville, we find this instructive modifier “deliberate”. In common speech we usually tend to use this adjective in the sense of doing something with intent. But Tocqueville and Lincoln (and Madison, as we’ll see) used it in the sense of doing something after discussion or consideration.

There is a difference between our everyday sense of the adjective “deliberate” and the republican one of Tocqueville and Lincoln. This difference can be seen when the law considers whether a murder was committed deliberately. To answer this, a court need only ask whether the murderer formed and carried out a bare intention to kill someone. If so, then the murder was committed deliberately. This is pretty much our everyday sense of the term. The court does not particularly care if the murderer considered whether, all things considered, it would be a good idea for him to kill someone. Republican deliberation, on the other hand, has to do with the wisdom of a course of action, and this kind of deliberation is best done in good time. As Tocqueville noted, this kind of deliberation “ripens slowly”. The deliberation of a murderer, by contrast, may take little time at all, and it usually has more to do with the best method of carrying out an intention than with the wisdom of doing so.

Modern US constitutional scholars tend to focus on the dangers of what might be called the “imperial Presidency”, the tendency for power and authority to become concentrated in the hands of the executive. There is good reason for this fear, as it is an observed tendency in modern US history. However, presidential overreach was arguably not the foremost concern back when the Constitution was being debated. For James Madison, perhaps the wisest of the Constitution’s framers, the greater danger lay in the legislative branch. Madison expressed his fears about the dangers of the democratically elected House of Representatives, worrying that it would become the tool of factions, the leaders of which would cobble together majorities to prey on the interests and property of the minority. His worst nightmare was an overly active legislature, because such activity would usually signify that it was up to no good. He saw the legislature as the branch of government most likely to encroach upon the jurisdiction of the other “members” of the Constitution, particularly because it would have the powers of popular support and the purse, which would constitute an unstoppable force against any “parchment barriers” standing in the way of its intentions:

“Will it be sufficient to mark, with precision, the boundaries of these departments in the constitution of the government, and to trust to these parchment barriers against the encroaching spirit of power? This is the security which appears to have been principally relied on by the compilers of most of the American constitutions. But experience assures us that the efficacy of the provision has been greatly overrated; and that some more adequate defense is indispensably necessary for the more feeble against the more powerful members of the government. The legislative department is every where extending the sphere of its activity and drawing all power into its impetuous vortex.” (Madison, Federalist No. 48)

As Madison saw, a fine balance needs to be struck. A democratic element is necessary in a constitution — to provide legitimacy to the government, if nothing else. The trick, then, was to find a way of slowing down that “impetuous vortex” of democratic legislative activity enough to keep it from undermining the state, allowing enough time for cooler heads to prevail. This is the essence of Madisonian republicanism: not creating checks and balances to stop legislation, but rather to introduce enough resistance to the system to buy time for sober second thought to reassert itself. As Larry Kramer puts it,

“Madison’s answer to the problem of republican politics had never been to create an external block on democratic decision-making. Nor had it been to remove the people from the process of governing. His solution was to complicate politics: to slow it down with internal checks, so that what ultimately prevailed was not the immediate reactions of an unreflective people, but rather a reasoned popular opinion that had been refined through a process of extended public debate. Each house of Congress, or the executive through its veto, could prevent proposed legislation from taking effect. But their block was really just a means to test the legislation’s merits and support by forcing advocates to respond to objections. The checking and balancing of the different departments of government thus served as devices to prolong the discussion of controversial proposals.” (Kramer, “Understanding Marbury v. Madison,” Proceedings of the American Philosophical Society 148 (2004), 14-26.)

Put another way, the purpose of checks and balances in the Madisonian republic is to enforce deliberation, so that self-determination of the people as their better selves may prevail. Unfortunately, the particular system of checks and balances devised for the US Constitution may have been too effective, for the American legislative process is often notoriously slow and prone to deadlock. On the other hand, it has displayed an admirable ability to rouse itself in times of imminent danger.

Meanwhile, Back in Canada…

Thanks to misguided British notions of the “legislative supremacy” of Parliament that it has inherited, Canada has historically had an overly strong “democratic element”. There was no effective constitutional countervailing power to slow down Parliament’s “impetuous vortex of legislation”. Fortunately, for reasons I have yet to properly understand, unlike in many other democracies, Canada’s legislative vortex has not been overly impetuous. There have been some ill-considered actions, but none that can be considered catastrophic. Probably this has less to do with any inherent wisdom in our constitution than with a natural conservative animus in the social and political culture of Canadians. Despite momentary fits of energy, Canada is a Red Tory nation, conservative in attitude and action if not in ideology. Although the nation’s Latin motto is A mari usque ad mare (“from sea to sea”), a more apt tag might be Festina lente (“Make haste slowly”).

If we once had to fear an “excess of democracy” in Canada, we do no longer. Quite apart from the unrepresentative nature of our first-past-the-post “democratic” legislature, we have been following the American trend towards an overmighty executive branch. However, unlike the US President, who still finds himself in the humbling position of having to please Congress, so long as he has majority control of the House of Commons, the Prime Minister of Canada can govern unhindered by anything other than displays of backbone from his cabinet members or the biting of his own conscience, two factors seemingly in very short supply.

Republicanism at its best represents a safe haven from the extremes of excessively democratic “mobocracy” on the one hand, and authoritarian tyranny on the other. It allows the time necessary for the democratic element of the constitution to get its deliberations right, and it provides the constitutional restraints necessary to prevent the dangerous gathering of power in the hands of an executive answerable to none.

Madison feared the impetuous vortex of democratic legislation. It was a fear well-founded. However, tyranny of the single despot is a greater danger to Canadians than tyranny of the majority at the moment. And it must be kept in mind that an emperor or dictator can be just as impetuous as a democratic legislature.

Canada’s dictator is a quiet man. He doesn’t like to discuss things. He shuns open debate. He has closed the Parliamentary press gallery (and has twice shut down Parliament itself). He avoids taking questions from journalists that aren’t softballs vetted by his staff beforehand. Thus, he seldom gives reasons for his actions, and when he does, they’re seldom good reasons, and he seldom gives opportunity for rebuttal. In this sense, nothing he does is deliberate (in the republican sense). As such, his actions are just as impetuous as Madison’s feared democratic legislature. However, his impetuosity stems not from an undue degree of passion — for he is seemingly passionless. Rather, he is impetuous because he acts only according to his own lights, which are unfailingly echoed by the flattering whispers of his hand-picked courtiers.

In this, he reminds one of poor Æthelred “the Unready”, the hapless Saxon king who acquired his nickname not primarily because he was unprepared, but because he didn’t listen to advice. (It was a pun: in Old English, Æthelred meant “noble counsel” and Unræd  meant “uncounselled”).

Friday, May 13, 2011

Publicans and Sinners

The Tax Collector
As a child sitting in church on Sunday mornings during one of those episodes when my parents temporarily found religion, I would occasionally hear Scriptural references to “publicans and sinners” (in Luke 15:1, for example). I remember my confusion about what a “publican” was. I had it in my head that it must be some sort of a tavern keeper or bartender. I plausibly associated it with the word “pub”. And I also associated such drinking establishments with “sinners”, not yet being old enough then to have discovered the peculiar charms of grape and grain. And interestingly, from around the 14th century, this really was the sense in which “publican” was used. And since back then, such keepers of public houses also often acted as pimps, well…

However, much later, when I became interested in Roman history, I found out that a publican was originally a tax collector, from the Latin publicanus. St. Paul was once a publican, before he had his vision on the road to Damascus. But learning about Roman publicani led me to a new mystery. You see, the publicani were not civil servants of the Roman government. Rather, they were private businessmen who had bid on contracts to collect taxes for the government. They would loan money to the government (at interest), and in exchange they would receive a franchise to collect taxes in a given area. In addition, they got to keep whatever they could extract over and above what they had to pass on to Rome. The publicani really came into prominence around the first century BC, as the Roman empire expanded and great profits were to be made in the tax farming business.

This system of public finance seems a very alien practice from our modern point of view, and it is not difficult to picture the abuses to which it would have given rise. Imagine that at tax time, a privately-contracted agent of the IRS or Revenue Canada knocks on your door (perhaps with a couple of large and threatening men beside him) and demands money. He demands not only what you owe the government, but also an extra “collection fee” to pay him for his trouble. Imagine that the only upper limit to his income is what he can extract from citizens like you. Even worse, imagine that — as sometimes happened — the government mistakenly (or corruptly) lets contracts to more than one such tax collector in your area. Now you have to cough up money to two people. Of course you could always appeal your case later, but that would take a lot of time and money, and Washington or Ottawa is a long way away, especially if there is no phone or internet.

In addition, publicans would often receive payment in kind rather than coin; they might undervalue the goods they receive from you and then sell them on at a profit. Tax collection abuse by publicani was the cause of many a provincial uprising. The Roman government knew exactly what these tax farmers were up to. It was expected that these publicans would be trying to earn their commissions; after all, they weren’t acting pro bono, out of a pristine sense of civic duty. The government only intervened if a publican overreached himself and caused a local disturbance that couldn’t be ignored by the powers that be. In any case, the publican was a byword for greed, which is why he is classed among the sinners in Scripture. He was, in effect, an extortionist.

But why would a government resort to a system that seems so inefficient? After all, the government would receive an initial loan from the publican that was less than the money available to be collected in taxes, and the loan would have to be paid back with interest. Why did it not occur to the Romans (and other tax-farming societies) to have taxes collected directly by government agents rather than by profit-taking middlemen? Well, for one thing, the Romans didn’t have highly developed economic theories, so they were only dimly aware that over-extraction by publicans might curb investment and economic growth, leading to a declining tax base in the long run.

But that alone doesn’t explain it. After all, the Romans weren’t completely stupid. They knew that the tax farmers were going to make their end somehow on the deal, otherwise they wouldn’t have bid on the contracts. And whatever profits the farmers made represented potential money that in theory could have, but never would, end up in Rome’s coffers. The publicani captured value that wasn’t being captured by the state. If we assume that all parties were self-interested, we must assume that the Roman government assessed that it was not cost-effective to catch the crumbs it knew were falling from its table.

A very old but classic paper by Nobel Prize-winning economist Ronald Coase entitled “The Nature of the Firm” (Economica 4 (1937), 386-405) can perhaps shed light on why Rome chose to rely on tax-farming rather than a public, centralized system of tax collection. Coase envisioned private market firms as being faced with a choice very similar to that of the Roman government’s: whether to internalize or externalize (i.e. contract out) one’s business activities.

Let’s say I come up with a product that I would like to produce and sell. There are two ways I might do this (with a large number of ways intermediate between these two along a spectrum). On the one hand, I might internalize all my production. For example, I might purchase the machines I need to build my product (or even design and build the machines myself), purchase the land and buildings in which to put the machines, and hire employees to work directly for me on-site in my new factory. I may also buy trucks to ship my product to customers. I may even decide to own the retail outlets that will market my products exclusively. In other words, I could have an ownership stake in every aspect of the process of production and marketing. This pure-type of internalized production is very rare: many businesses will, for example, rent rather than own their production space, which is a kind of contracting out. As I said, there is a spectrum of possibilities.

At the other end of the spectrum, I may decide to contract out just about all of these processes to other parties to do on my behalf. For example, I might pay someone else with machines and factory space to produce the goods for me, with workers to be hired by them. I may rent trucks rather than own them, or simply hire a trucking company to ship goods for me, thereby not having to hire drivers either. I may be the only person directly employed by my firm. Again, this is an imaginary pure-type: unless it were a very simple and small business (e.g. selling homemade pet rocks) it’s hard to imagine a scenario where I could externalize literally the entire process. But in theory at least, much of it could be.

The point is, neither internalized nor externalized production is a priori more natural than the other. So instead of being surprised by the choice of the Romans to externalize much of their public finance, we should rather ask why they chose that route rather than a more internalized bureaucracy.

According to Coase, if the market were perfectly efficient there would never be such a thing as a firm, because in an efficient market it would always be easier to contract out. However, the market is never perfectly efficient. There are always transaction costs to bargaining, costs such as time spent in procurement, in seeking and communicating information, in bargaining, in enforcement of contracts, etc. There are also risks, that other parties won’t perform their end of a contract, or will make mistakes, or will divulge trade secrets. Many of these transaction costs can be avoided or minimized by internalizing them within a firm. It then becomes easier, for example, to monitor workers, to control the flow of information, or maintain standards of quality.

Generally speaking firms tend to become larger — more of their production relationships are “contracted in” — when such contracting in will lower transaction costs and/or stabilize them, or when doing so will reduce the likelihood of production mistakes, or will lower the cost or raise the supply of factors of production.

However, there are limits to the process, and it can move in the other direction, towards greater contracting out. This might happen, for example, where overhead costs increase, or where a firm becomes too big or its geographical area of operations too widespread for effective oversight or decision-making, leading to mistakes and poor resource allocation.

Returning to the Roman situation, in the Republic and early Imperial periods, tax collection and other services were contracted out to publicani. It may simply be that in early times, the Romans were culturally averse to centralized government. By later standards the early Romans had a relatively weak state. Even firefighting was privately run, and there wasn’t even a public prosecutor. Later, there was an increased movement towards internalization, with many tasks that were once performed by government contractors being performed by a growing imperial bureaucracy. With regard to tax collection, perhaps part of this process was driven by the government’s inability to effectively control the “entrepreneurial” methods of the publicani. The mismanagement of the latter would have added significant costs to the government in the form of military resources deployed to quell provincial tax revolts. As expensive as the imperial bureaucracy was, it may have become (for a time at least) more cost effective than the system of publicani, especially at first, when the profits accruing to the publicani began to accrue to the government instead.

In our time, contracting out rather than in seems to be the general trend in both the public and private sectors. For example, my city has started to contract out its garbage collection. And in 2006 the IRS started contracting out its taxpayer debt collection to private debt collection services — the new publicani of our age. Presumably, with the salaries and benefits of government employees increasing faster than those in the private sector, a point is reached where contracting out results in significant savings for government agencies like the IRS. If history is any lesson, perhaps we can expect stories of tax collection harassment and abuse to become more prominent?

As a side note, Ron Paul has announced that he will seek the presidency in 2012. Since he is on record as advocating the abolition of the IRS altogether, perhaps under his administration the government of the United States would completely revert to a tax farming system. Vanitas vanitatum, there truly is no new thing under the sun.

Tuesday, May 3, 2011

Insights from Insite

For someone like me, who is of mildly conservative leanings, the application of public policy can be visualized as the dropping of a stone into a calm pond. Much like ripples, public policy has social effects that radiate outwards from the centre of social action. The most proximate effects are the ones that tend to get noticed, these being more pronounced and closer to the point of impact. The concentric circles further out are less pronounced and eventually fade away into imperceptibility.

However, while those inner ripples are fewer in number, the outer ones are more numerous and widespread. Thus, like ripples, the greater overall effects of a policy might just be those that we least take notice of. Herein lies the task of the conservative social policy analyst. Where others might be tempted to focus only on proximate effects, she must look to the long-term and often indirect effects, to that which is hidden. She must do this not out of a reflexive impulse to pooh-pooh any measure that smells like change. Rather, she must do it out of a desire to urge us to slow down, just a bit, and to consider for a moment whether we may not be creating unintended consequences through the best-laid plans of mice and men.

For some time now, there has been a social experiment going on in Vancouver’s drug-blighted Downtown East Side. It is called Insite, a program that provides a safe injection site for junkies to shoot up in. The main aim of Insite is to cut down on the rate of drug-related deaths by users. A recently-released study indicates that, by this standard, the program is a clear success. There is now talk of expanding the program to various other cities.

However, the current federal government here in Canada has always hated the idea of Insite and has bent over backwards to try to get it eliminated, to no avail so far. They don’t like the idea of coddling a group of citizens they consider to be little better than criminals, and they have gone to some lengths to provide their own counter-evidence purporting to show that Insite is actually a failure.

The government is on firmer ground when it appeals to an economic justice argument, for the fact is, Insite technically subsidizes junkies to shoot up, using funds presumably provided by taxpayers. If a convincing case can be made that taxpayers don’t want to subsidize junkies, then this to me represents a compelling case not to do so. On the other hand, if the program works, and if taxpayers can be convinced that it’s worthwhile, then, Insite is economically defensible. And if Insite relied only on charitable donations rather than tax funds, then no economic injustice would be committed. But I have a gut feeling that the current government would still hate and oppose the program in either case, so their aversion to it really has little to do with economic justice as such

I loathe the current federal government, so it pains me to find myself in ― qualified ― agreement with them. I have grave reservations about Insite. However, my reservations are not based on the federal government’s preconception that addicts are ipso facto criminals rather than persons. Thus, in what follows I will address their problems as if they were upright citizens deserving of just as much respect as you or me, albeit they happen to be citizens with a very serious health problem.

Also, my grave reservations about Insite are not based on any skepticism about the validity of the study indicating Insite’s success. I am willing to grant that the available studies are correct: Insite has been successful in its stated goal of reducing drug-related deaths in Vancouver’s Downtown East Side.

However, the successes focused on in these studies are what I was referring to above as the inner ripples of policy, its immediate or proximate effects. But what about the outer ripples, the ones that are harder to see or to measure? I heard an interview on the radio last week with a senior administrator of the program (unfortunately I didn’t catch his name). He stated that the goal of Insite was to “keep addicts disease-free until they are ready to make the decision to seek treatment for their addictions” (I’m necessarily paraphrasing here, but pretty accurately, I think).

The evidence indicates they have indeed been kept (more) disease free. This I won’t dispute. But it begs a larger question: What warrant do we have to assume that untreated addicts reliably make the decision to seek treatment? Evidence here varies, but most of what I’ve seen is not encouraging. That is one of the unseen ripples of the policy.

We can also follow the ripples a little further out. For example, what warrant do we have to assume that untreated addicts reliably make the decision to seek treatment when they receive services that enable them to continue using drugs relatively safely? If Insite facilitates addicts’ safe drug use, might it not be the case that a very strong incentive to quit ― the high probability of disease or overdose ― has been removed? This question was screaming to be answered in the interview, but neither interviewer nor interviewee seemed much interested in addressing it. On the interviewer’s part I imagine this was motivated by an implicit desire to display his social liberal credentials to a mostly like-minded CBC audience.

Defenders of Insite have argued that the availability of information and professional consultation available at Insite will help guide users to treatment options. Perhaps this is the case, but again, it is an empirical claim for which we want evidence. The report touting Insite’s success simply tells us that addicts using it are more likely to remain disease-free. It is (tellingly?) silent about how many of them end up seeking treatment. Until we have information on this, the claim that addicts who avail themselves of Insite’s services are more likely to seek treatment remains an unproven hypothesis, not a fact.

There is further reason to doubt this claim’s validity: most professionals in the “addiction services industry” (I don’t really know what else to call it) will tell you that there is a shortage of spaces available in treatment facilities. There are too few spots to service current demand, let alone an increase in demand.  Logically, this means that it would be impossible to treat more patients given current resources. So Insite cannot be leading more addicts to treatment, because the treatment isn’t there for them to be led to.

Obviously this is not an argument against Insite’s claim to efficacy in bringing addicts closer to treatment. If anything, it’s an argument for an increase in resources devoted to addiction treatment. Again, however, we must consider the policy ripples a little further out. Might it be the case that the increased access to addiction treatment information is at least somewhat ― if not entirely ― offset by the convenient and comfortable drug injection environment offered by Insite? At the very least, we must admit that mixed messages are being sent here: “We’d like you to consider getting treatment for your addiction. Here’s a brochure. And here’s a clean needle and a comfy room to shoot up in.” A junkie could be forgiven for taking away from this the message that his addiction is not so bad for him after all, and that treatment is not imperative.

Much of one’s attitude toward programs like Insite will depend on the views one has about the nature of addiction itself. Addiction is a puzzling phenomenon. Once upon a time, it was simply viewed as a matter of lack of willpower, and was thus regarded as a moral problem, a moral failing. We could call this the “moral disease” view of addiction. It has to a large extent fallen out of favour among researchers (along with many other things that employ the term “moral”). Speaking very broadly, the moral disease view has largely been replaced by either of two currently popular models of addiction. Fortunately, my reservations about Insite are unaffected by whichever of these views one happens to hold.

The Disease Model

Probably the most widely accepted and fashionable paradigm for understanding addiction is the disease view, which treats it as if it were a medical condition on all fours with, say, diabetes or cancer. Of course, there is a large psychological component to this particular disease which differentiates it from purely physical conditions, but for treatment purposes, addiction on this model is seen as a medical condition, a psycho-physical sickness beyond the control of its victim. This is the operational viewpoint of most Twelve Step programs.

From what little I know about the history of addiction treatment (and it’s probably more than the average person, for personal reasons I don’t wish to expand on here), the disease model was developed in the treatment of alcoholics in the early 20th century for at least two reasons. One was that clinicians noted that chronic alcoholism had a more or less definite disease pathology, a natural progression or history of symptoms, when left untreated. Another was the hope that by seeing alcoholism as a medical condition, people would be less judgmental of alcoholics, and alcoholics would be more willing to seek treatment once the stigma of their condition was removed. In other words, it was specifically hoped that the disease model would replace the old moral disease view with its accompanying social stigma.

Both of these were valid points to some extent. Chronic and very acute alcoholism does have a fairly well-defined physical pathology. And probably more problem drinkers have ended up seeking treatment than they otherwise would have once the social stigma attached to the “disease” was mitigated.

On the other hand, not all problem drinkers fit a disease profile that was originally developed with the most advanced late-stage alcoholics in mind. It was a one-size-fits-all model that has not proved very helpful in understanding the “disease”, if disease it be. And although more people sought treatment, the treatment programs based upon this disease model have been notable only for their abysmal success rates. Alcoholics Anonymous has been around since 1939. It does not track its success rate in treating alcoholics. You’d think they would want to. But there is probably a good reason why they don’t. Others have done the studies on their behalf, and the general consensus is that Alcoholics Anonymous is a failure at keeping people sober. There are far more failures than successes, and even among the “successes”, relapse rates are sky high. If alcoholism is a disease, and if A.A. is the “cure”, then as a doctor, I’d consider taking my chances by prescribing a placebo to my alcoholic patients.

Long-time members of A.A. will counter by saying that the organization is not supposed to “keep people sober”. Alcoholics are supposed to keep themselves sober; the responsibility for recovery always lies with the alcoholic herself. But doesn’t this seem in tension with the notion that addiction is a disease for which its sufferer is not responsible? A.A. members will also tell you that failure or relapse happens because an alcoholic can’t be treated unless he “is ready” to recover, or has hit bottom and decided to give up the drink. But again, this seems in tension with the disease model’s stress upon the notion that alcoholism is not a function of the alcoholic’s lack of willpower.

Sooner or later it seems, every treatment scheme predicated on the disease model runs up against this tension. The disease model may be better at luring the addict into treatment, but at bottom, the treatments themselves end up stressing willpower and the addict’s own inner resources for recovery. Alcoholics Anonymous is no exception. As a matter of fact, their entire program of recovery can be viewed as a conscious attempt to instill in the alcoholic a transformational revolution of character, the beneficial moral effects of which will hopefully mimic those of intense religious conversion. It is no coincidence that Bill Wilson, co-founder of A.A., was greatly influenced by William James’ The Varieties of Religious Experience. It is important to stress that such an approach works on the internal resources of the alcoholic’s own moral agency, which seems rather at odds with the disease model of addiction. After all, we don’t cure cancer or diabetes by relying on the character or willpower of the patient.

In a true disease model, the patient is just that ― a patient. But addiction treatment almost always ends up (as it must, I believe) treating the patient as an agent. Perhaps the success of such programs might be improved if they were to be honest about this from the beginning. Maybe we must fess up and admit that kicking addiction is a matter of willpower after all (helped along, of course, by a certain degree of moral support from others). Obviously, putting it this way grossly oversimplifies the phenomenon of addiction. But then again, so does the disease model.

The Rational Choice Model

The other school of thought ― the rational choice model ― is less popular than the disease model, at least among clinicians. The rational choice model has been adopted mainly by behavioural economists because it fits well with the observed behaviour of addicts. In a nutshell, this model views the addict as a rational consumer, in almost all respects the same as you or me.

Generally speaking, when the price of a good increases, consumption of that good tends to fall. The rate of decrease of demand will vary from one good to another, and from one agent to another, but the general trend is obvious. The addict’s consumption of drugs is, in principle, no different. If you lower the price of cocaine or heroin, addicts will tend to increase their consumption of them, and some people will begin to consume these drugs who otherwise might not have.

The fact that the addict is addicted to the drug simply means that his demand for the drug is inelastic relative to other goods. If the price of heroin goes up, before he considers cutting down his consumption, the junkie will instead often forego food or rent. This seems like an irrational consumption choice, but if each of us were to systematically examine our own consumption choices, we would often find similar though perhaps less harmful patterns. I once knew someone “addicted” to electronic gadgets; he absolutely had to have the latest thing, spending well beyond his slender means, even if paying for it meant that he didn’t know how he would make his rent or pay off his ridiculous credit card bill. My wife might say that my penchant for book collecting meets this description. There just happens to be no self-help group akin to Alcoholics Anonymous that I know of for my particular form of consumption aberration (if there were, I’m sure my wife would have left a brochure on my pillow by now).

The rational choice model views the addict as making choices about consumption, based on what they believe will best satisfy their overall preferences. To the third party observer, these choices can seem downright irrational, but to the person making the choice it seems perfectly rational: they have a (very strong) desire for a drug, which gives them pleasure or ― what amounts to much the same thing ― relieves their pain. Any approach to treating the addict should treat them as rational persons making choices based on a preference set and an information set. Such an approach would work on improving their decision-making by modifying their preferences and providing them with information. Inevitably, there will also be situational and environmental factors standing in the way of better decision-making, things such as easy availability of the drug, hanging out with the wrong crowd, etc.

The point is, rather than approaching the addict as a powerless patient whose addiction is an external force majeur that strikes him through no fault of his own (the disease model), the rational choice model treats the addict as an empowered, decision-making agent, responsible for his own conduct. He ought to be praised for his good choices, especially when, as in the case of addiction, good choices are difficult choices. And he ought to be held responsible for his bad ones. The prevailing view that addiction seizes and controls the addict, as well-intentioned as it might be, is probably misguided. It is one part of an overall tendency to “medicalize” conduct that is moral in nature. I will illustrate this medicalization of moral conduct with an anecdote from an essay by Theodore Dalrymple.

Dalrymple was a prison psychiatrist. He was interviewing an inmate who was in prison for viciously beating up his girlfriend badly enough to put her in the hospital. This was not unusual behaviour for this particular inmate. Dalrymple asked him why he beat up his girlfriend. He replied to the effect that he had difficulty controlling his anger and she had made him really angry. Not really his fault, you see. The general gist of his answer was that he was in prison due to a problem largely beyond his control ― he had “anger management issues”, to use the parlance of our times. He simply couldn’t help beating his girlfriend to a bloody pulp. Dalrymple then asked him how it was that since he had been in prison he had managed to be well-behaved, with no outbursts of anger or violence. He replied that the guards didn’t let prisoners get away with that kind of bullshit in prison.

The prisoner had ascribed his violent behaviour to a quasi-disease model of anger, which he probably picked up by osmosis from the liberal claptrap he was hearing around him. Working from within the disease model, the prisoner’s subsequent good conduct behind bars makes no sense. But working from the rational choice model, it makes perfect sense. He was able to modify his behaviour in the presence of a strong incentive, namely getting the proverbial tar beaten out of him by a prison guard. Research on addiction done by behavioural economists bears this phenomenon out.

This has obvious implications for programs like Insite. When we ask ourselves what motivation Insite offers addicts to quit, we mostly come up empty-handed. Again, yes there are brochures and addiction counselors, but there are also the clean needles and the safe and comfy shooting galleries that cancel these out. The rational choice model predicts a poor long-term treatment prognosis for users of Insite. I hope someone will do the research to test this hypothesis. It represents one of those outlying ripples of this particular social policy, one of those rare ones that is, in principle, empirically measurable. And the test should probably happen before the program is expanded, not afterward. That is how conservative policy analysis should work.

UPDATE:
At time of writing this addendum (April 28, 2016), almost exactly five years after this post was originally published, there is still no word about whether Insite is successful at getting more addicts into addiction treatment. I must assume it has not been successful, since its vociferous "progressive" supporters never hesitate to trumpet its other supposed "successes". Hence, I must regard their silence on this point as eloquent.

Germane to my comments about the longer-term "outer ripples" of public policy with which conservative policy makers ought to concern themselves, it seems that, five years out, one of those mid- to long-term unintended consequences is making itself apparent, namely the hundreds of thousands of used needles turning up in schoolyards, children's playgrounds,  and residents' flower beds in Vancouver.