Monthly Archives: January 2012

On a random peregrination of the blogosphere, I stumbled on this at

The modern world depends on economic growth to function properly. And throughout the living memory of every human on earth today, technology has continually developed to extract more and more raw material from the environment to power that growth.

While the rest of the article is of some interest, this premise is a sadly very common fallacy. The first sentence is pretty debatable, but for today we will assume it true and concentrate on the implication of the second, that economic growth needs stuff.

At the dawn of the industrial revolution, a couple of centuries ago, you probably would have found people arguing that the economy could not grow much anymore, because all the land that was suitable for agriculture was already in use (in Europe) and demand was close to being fully satisfied, so there was no room for growth. They were even right: not only the land cultivated stayed more or less constant, but technological progress allowed the amount of human effort expended on growing food to be considerably reduced, to the current state of affairs, where, in developed countries, 2-3% of the workforce is all that is needed to feed everybody. This then allowed people freed from subsistence farming to move on to other activities, first via to the manufacture of objects, and then, in the West at first, to service industries.

Industry is the next agriculture: it’s becoming so efficient that we will soon probably need less than 10% of the workforce to manufacture every object desirable, within reason. The globalisation steps which saw manufacturing move to East Asia thanks to low labour cost will just be a parenthesis. As their labour costs increase during the normal course of development, they won’t even bother finding even cheaper countries to use the cheap labour of, but skip direct to further automation, which has great scope to be further implemented in a world where cheap labour has oft been chosen in the past in place of then more expensive automation. This applies even if the technology remained static, as the current state of the art of automated manufacturing allows to manufacture almost anything, it’s just a question of getting enough volume to justify the relatively steep capital outlay.

Politicians who care where widget are made are clearly living in the wrong century. Wherever it is done, it ultimately won’t keep many people busy. Only maintenance and monitoring staff are required once you’ve installed the machines built by other machines. This is a very good thing, the same as humanity freed itself from the burden of having to spend most of its time to feed itself, it can now free itself from the burden of having to manufacture widgets.

Once freed from that burden, people can do other things. They could choose to work less, as it requires less time to produce a constant material standard of living, but they can also choose to keep on being economically employed if they so wish, by moving to more abstract activities, that are not resource intensive. This can already be observed by looking at the difference in consumption between the middling sort and the proper well off in the West. The wealthy don’t eat more than trailer trash, indeed they probably eat a bit less in raw quantity. They may spend more money on food, but little of the difference will go to the resource intensive bit: a posh restaurant delivers approximately the same quantity of food as a fast food joint, it just deliver it in more style: you pay for the waiter, the cook spending more time on a sophisticated arrangement, the interior designer who chose the curtains, the marketing people who designed and sold the concept, etc. It’s more entertainment than catering: most of the staff involved in the restaurant experience are performers whose task is relatively light in the use of resource. Such freedom from materiality applies everywhere: for instance the wealthy do spend more on divorce: they employ lawyers and judges who build complex narratives on top of the simple story of a couple splitting up. The courthouse is also a form of entertainment: it’s where people tell stories to each other, like the theatre. And they can do that because they’ve freed themselves from having to hunt for food (or widgets) all day — but it’s a choice, they don’t have to, poorer people still do divorce sans decorum and probably enjoy the experience almost as much.

Back to economic growth, the move to abstract activities like all these forms of entertainment can expand as far as people wish them too. Humanity can easily keep itself busy all day with that, and other low-resource things like scientific research, and it can do that within the economic sphere if it so wishes. There is simply no dependency on raw materials for growth. The only constraint is a bit of imagination.

Besides the resource usage problem contains an intrinsically stable virtuous cycle: if there is a shortage of materials, materials-based activities become less desirable and so people will naturally move to less resource intensive alternatives, which become, relatively, more compelling. The opportunity cost of resource-based activities increase with every increment in the price of energy of raw materials. People don’t go on a two week holiday on the other side of the planet because they need to but because, for a moderately well-off Westerner at the current juncture, they can. The amount of extra enjoyment you get per kilometre compared to going to a similar location on one’s own continent is pretty minimal. If the fuel necessary for a return flight to the other side of the planet costs say a couple of years worth of that moderately well-off person’s income, they’ll think twice before going that far for a couple of weeks, but will surely replace it with a less resource intensive substitute, say a train trip, or an intensive yoga course, at probably very little cost in therm of loss of practical enjoyment, and causing no impairment to economic growth as human labour is just redirected towards new activities other than digging up stuff from the ground for the purpose of moving around just for the sake of it.


I got irked by one of Bill Mitchell’s latest, where he claims:

A reliance on credit-driven consumption is not sustainable overall.

This is a well accepted orthodoxy, but is it really the case?

Let’s look at how consumer credit operates. To avoid introducing irrelevant systemic issues that are not salient to the main point, let’s suppose that all consumer credit is issued by specialised operation who raise the funds they lend to consumers from the stock market via share and bond issues.

Now what happens if enough consumers default to take (some of) such organisations into bankruptcy? Nothing special. People putting money into ventures, some of whom are not successful, is the basic function of open markets. Who, then, pays for these failures?

  1. Consumers who do not pay back their borrowing and go bankrupt and will, for a little while, “pay” in the form of losing access to credit, though they do benefit from what they spent before bankruptcy (more on that below)
  2. Consumers who are in the same risk cohort as those who go bankrupt but do pay back,  via inflated interest rate
  3. Investors in failing consumer credit companies

All these are transfers within the private sector that do not change public/private sectoral balances. But do they impact aggregate demand?

  • Transfers from #2 (diligent borrowers) to #1 (dilettantes) should be neutral as both groups are net borrowers with a maxed out propensity to spend
  • Transfers from #3 are directly a conversion of savings into consumer demand, and it’s unlikely any other form of savings would indirectly cause more demand, so they seem clearly positive for aggregate demand

How is this sustainable in the long term? In efficient markets, investors should learn and stop losing to insolvent consumers, but there seems to be heaps of evidence that people are attracted to high risk by illusory high returns — it’s a corollary of the high returns from low volatility anomaly. As long as this remains true, the transfers from these investors willing to lose money to consumers is totally sustainable, and even desirable from a social justice viewpoint as it’s a way to realise transfers from the well off to the less well off.

Distressed but diligent borrowers are not so clearly benefiting, though the extra cost may allow them to do things they couldn’t do otherwise if completely locked out from borrowing; but in any case there is no doubt this category will remain in existence.

So, there doesn’t seem to be any reason for policy-makers to try and restrain exuberant consumer credit when they want to support aggregate demand. It seems overall socially beneficial, and, within the limits of the supply of agents willing to lose money to the operation, totally sustainable.


There is an interesting side question: does it make sense to be on the “dilletante” side of the deal, and as a consumer borrow when you know you’re unlikely to be able to pay the loan off? For people with positive net assets, or merely assets larger than the scale of consumer loans, clearly not, as they have something to lose of greater value than typically modest consumer loans.

For people who have virtually no assets, if they can borrow, even at relatively high interest rates, it may make sense, at least in jurisdictions where bankruptcy proceedings are not too taxing. Spending a few years without credit might be a mild price to pay for some years of extravagance. The important thing is to spend the proceeds of the loans on things that the creditors can’t seize, such as education (you keep the skills), travel, or edible substances (you keep the memory of the good times).

It could be argued that if it becomes socially acceptable for relatively poor people to do a runner on loans, social cohesion will be reduced. This is entirely correct. That said, should poor members of society be expected to have higher moral standards than financial industry high fliers, some of whom have certainly done things that are much less socially cohesive than spending a few thousands on a holiday and not pay it back?

The main proponents of MMT (Modern Monetary Theory), like  Bill Michell and Warren Mosler, have been lately pushing further their “Job Guarantee” proposals, that of a job paid by the state, something in the vicinity of the minimum wage, for anyone “willing to work” who can’t find a better paid job in the regular private or public sector.

I have no objection on the monetary policy side of things (it’s consistent with the rest of MMT) and not even in principle if we lived in perfect societies where everyone unemployed is willing and able to work and can be allocated to a useful but previous uneconomic task friction-less, then it could possibly even work.

Unfortunately the real world is not like that. The Job Guarantee is not a new idea. It’s not even an idea, it has been done in practice, particularly in Europe, under various schemes which are operationally very similar, even if they were not universally applied and not called job guarantees, but they give a lot of material that can be useful to extrapolate the difficulties a full scale scheme would encounter.

At the very basic level a job guarantee is strictly speaking a logical oxymoron: a job cannot be a job if it’s guaranteed. The proof is simple. Let’s imagine we have an unemployed individual person with no desire to work, but with a desire to get the wages from the job guarantee (if they are above non-working welfare benefits, which is what the MMT proponents want as far as I know). How would this person operate? Probably thus:

  1. Go to a job guarantee office to get allocated a place
  2. Turn up but refuse to do anything
  3. If “fired”, go back to #1

So it is not a job as you don’t have to work. It’s just daily detention for some welfare claimants. There is no escape from this: either you remove the job part (people can just turn up to collect the money) or you remove the guarantee part (people who don’t work go back to regular unemployment, so they are not guaranteed a job). Which brings us to this invariant:

You can have a job that’s not guaranteed, or a guaranteed income that’s not a job, but not both at the same time.

It could be argured that the job guarantee only applies to people “willing to work”, and that those who aren’t are simply disciplined out of the system.

The problem is that, if you assume that people running the work scheme are trying to get something useful done, they will manage their workers the same way as regular employers do and fire, or refuse to employ in the first place, the people who are not at least marginally productive. This will also include the objectively willing being useless and those who could be useful but are not willing, because you cannot operationally distinguish them: a good actor pretending to be unable will be virtually impossible to distinguish from someone who is genuinely useless. That brings us to our second invariant:

Willingness to work is not an observable characteristic (in a non-totalitarian state)

The only way I can think of to resolve that problem is to organise the job guarantee in the form of prison work, that is a coercive labour camp, or the Victorian workhouse, which even the Victorians finally had to admit was impractical even when armed with a diabolically repressive moral code. This is not acceptable in a modern civilised state (even non-democratic ones).

Picture of 19th century workhouse

A Job Guarantee centre?

In fact it could be said that the entire unemployment problem is, at all times, an issue of unwillingness to work at the margin. We could probably employ all the unemployed in the world as garden gnomes paid $1 a month, and there would probably enough demand for live garden ornaments for next to nothing. Virtually all able bodied unemployed could do this task. Most would, rightly, be unwilling to at this price.

When pressed, some commentators in favour of the scheme will water it down to some support for say charity workers being paid by the state. This is not a job guarantee as charities have objectives to fulfil and will only employ workers that have positive productivity, they have no need for people who are simply useless, or negatively productive (require surveillance to prevent them from doing damage, or extra work to undo the damage they cause through ineptitude or inadequation to the tasks at hand). Therefore, unless coerced into becoming labour camps, charity or other voluntary sector organisations will operate like the paid-for economy does: as a market which matches demand and supply, some of which, unemployment, doesn’t clear.

Reduced to that form, it could still be useful, but the description is misleading: not only this is not a job guarantee, but it does not need to be introduced as a grand new scheme. The government is already tasked with doing things that are uneconomic, and it should be enough to simply add to the credit of local or national government departments, so that they employ people the usual way (temporary jobs for temporary credits, which might include paying voluntary sector organisations’ workers if that’s the most expedient way to achieve public good objectives). The level of subsidy can be tweaked until the unemployment level goes down, within the limits imposed by the requirement that the market for jobs still clears, that is (almost) all credits being spent and not saved due to the lack of applicants.

Once we’ve accepted the job guarantee is not guaranteed, we’re left with a temporary increase in government expenditure, and there’s no reason to invent a nomenclature and separate bureaucracy for something that’s already part and parcel of government operations.

Not only would this be silly, but it’s a massive marketing error from MMT proponents. If the job guarantee is about matching government expenditure with the unemployment situation, it’s acceptable to many constituencies in the spectrum of public economic opinion, while presented as a “job guarantee” scheme that cannot possibly work in reality and is very naturally mistaken for unacceptable things, it’s unacceptable to virtually everyone except a fringe. This might induce many of them to reject the whole of MMT en bloc, for no good reason as the rest of MMT is solidly grounded in common sense.

The probable bankruptcy of Kodak is all over the web, with everybody apparently concluding they failed, as a technology driven company, to adapt to the digital era.

From a cursory look at their product portfolio, it seems quite okay, well diversified with lots of speciality products in niche markets, especially on the professional side, most of which are probably stable businesses. The consumer products may not often be class leaders but seems to have some unique qualities, seems middling but not obviously failing.

The surprise came when I had a look at the balance sheet. It’s limpid. A company with 6 billions of assets has a 2.5 billions unfunded pension liability, gross, all of which nets to 1 billion negative equity. So it seems there’s a perfectly fine and viable business were it not for the legacy pension issue. The lesson is the rather obvious: a large business can’t scale down in size when it has an open ended pension liability that, as here, might prove underfunded, at least from an accounting or regulatory viewpoint, if the market or regulations go against the supporting portfolio at an inconvenient time.

So a GM-like restructuring or outright dismantlement seems necessary, hard to see how shareholders (for what little equity there is left) or bondholders could get something back, but once the process of cleaning them out is done, I suspect a large part of the product lines will survive, either sold off separately, or together.

The question is why so few people are mentioning that — I must not be alone but it’s silly that I had to rediscover this independently. The summary balance sheet on financial websites may not itemise the “pension benefits – underfunded” line, but it doesn’t take much to find out.

Some time ago deep in a comment in some forgotten thread, Résultat d’Exploitation was giving salt as an example of a simple product with little scope for price or product differentiation.

I thought it was a bad example given that there’s ample scope to market hand collected, sun-dried or otherwise carefully brought up, salt with “unique” taste or qualities, from the Mediterranean or the Himalayas, or a salt mine on the moon. Given that the salt budget is going to be negligible in most people’s budget, a sizeable proportion of the potential customers should be quite price insensitive. I expected an order of magnitude variation between entry level and premium products.

I’ve now checked in my local, mid-range, supermarket and I still managed to be surprised. Of about a dozen table salt products available, the cheapest plain cardboard box of NaCl costs 38 euro cents per kilo. The most expensive one, in a fancy jar with some added background taste additive, surely hand collected by Ph.Ds in saline solutions, costs… 42 euros per kilo, and there are several products which are not much cheaper. That’s a factor of 110! Makes you wonder how much premium food is just ordinary food repackaged with a fancy narrative. It must be tempting.