We did Martin's tax. He is going to get a $3 tax return. He is pretty angry and screaming a little bit.

Some people have accused the IMF of managing a "conspiracy" of banks and wealthy nations, to prevent poor countries from declaring some of their debt as unfair and refusing to pay it back. The poor nations claim that the money was borrowed by dictators who mostly kept it for themselves or used it to kill their own citizens. It seems like a reasonable accusation because the IMF proudly takes the responsibility for creating creditor cartels and its mandate is to preserve financial stability around the whole world, not to help poor countries get less poor.

However, I've found an even more persuasive sign that there's a conspiracy. And it's the piss weak explanations that come out of the IMF attempting to explain why there's no conspiracy. The argument against letting countries not pay back unfair debt is essentially this. If a government is labelled as corrupt because it's stealing the money it borrows, then banks will stop lending money to it. And if banks stop lending money to it the government won't have any money with which to develop the nation.

Seriously. That is all they come up with. It strikes me that if you decide a government is corrupt and stealing development money, then you'd be happy that banks stopped lending to it.

The author sets up a couple of straw proposals so he can fluff out the article explaining what is wrong with them. But the actual model that is proposed by other academics, he only has one argument against. And it's total bollocks. He lays out this one-paragraph argument, and then smugly say there's obviously no need to resort to conspiracy theories - odious debt proposals are apparently going to stay in "cold storage". On the contrary, there seems to a real need for some other explanation. And if a conspiracy theory is the only explanation anyone can find, then why not run with it.

Besides it isn't much of a conspiracy theory. The IMF itself only offers one sensible reason for forcing poor countries to pay back their debt - that the rich countries can't afford for them not to.

I wonder if anyone has considered setting up derivatives markets for famine. You could have a futures product (or many) linked to some aspect of the calorie intake for a particular country or area. It could be the mean (although probably not) calorie intake, although including the variance would be better. It could be the proportion of the country consuming less than some figure or just the fifth percentile of the distribution.

You probably wouldn't have starving people going into the stock exchange buying futures in their own calorie intake variation, but insurance companies could use the markets to fund some sort of insurance product. If the price of the insurance wasn't extremely cheap, that would be a sign that the government or the UNDP or whatever needed to grow some food quick smart.

Of course, an insurance company could just offer insurance the conventional way but given that famines tend to affect large numbers of people at once it might not work that well.

Using a futures market might make it easier to reduce moral hazard and adverse selection problems. If one person in a village was starving and made an insurance claim while everyone else was doing fine, the person making the claim would have to show how they were affected by something that no one else was.

Collecting data on calorie intake would be tricky, because there'd be money at stake based on the measurements. People might bribe doctors or whoever to change the data. But if it was collected by some UN agency (which I guess they already do), then it would probably be OK.

There's also the problem that people in positions of power (bureaucrats etc.) might be able to profit by causing famines. But selfishness probably already causes quite a few famines, so perhaps having one more avenue wouldn't be such significant thing. There are already plenty of examples of functional derivatives markets where people could do deliberate damage to governments or companies in order to make money. Perhaps that is even implicit to derivatives markets. As soon as people were able to profit from things going wrong in the world - which is largely what makes derivatives special - there was a greater incentive for sabotage. I'm not sure if famines are sufficiently different that this would be a real problem where it hasn't been to date.

Probably a lot of people will think this idea is pretty sick...

...which I've just realised adds another interesting aspect. There's the potential for the sickness factor to "distort" the market. If speculators are irrationally biased against selling short famine futures they'll prefer to bet on food intake going up. Meaning, if they don't want to make a whole lot of money out of people starving to death then the market will be biased in favour of the starving, who are effectively selling short their own nutrition levels. Which is a slightly strange concept.

It seems that when heavily leveraged investors go bust the standard excuse is always "But markets aren't suppose to do that." The whole point of leverage is that when markets are behaving "normally" you can make a bunch of money, but when they behave erratically you can lose all the money you made before you know it. When Long Term Capital Management went bankrupt they used this excuse. If only returns had just stuck to their harmless normalishly distributed movements, everything would have been fine. I've heard an equivalent complaint from one firm damaged by the recent movement in US credit markets.

If leveraging didn't require a very difficult skill, then you wouldn't be able to make so much money out of it. If you're going to use heavily leveraged investments over the long-term you'd better know when to unwind, because sooner or later you will need to. It's no good whingeing after it's all over that you were too stupid to know what your risk exposure was.

The only thing worse than a whinger is a rich whinger who has everything going for them except their own competence.

My first and hardest exam is over. It wasn't actually hard at all. I am very glad.

I like this graph. The yuan is totally the Chinese government's bitch.

I've had a look at a few different things for stuffing around with matrix arithmetic. It's the sort of thing my degree has reduced me to. I've looked at Excel and R. I already know about Stata. At uni they think Matlab is pretty good. But I've found something better than all of them. Ruby. It's totally sweet.

require 'matrix'
Q = Matrix[[20, 30, 43],
           [20, 25, 10],
           [20, 10, 5]]
assets = Matrix[[10], [20]]
price = Matrix[[5, 10]]
portfolio = Q.inv * assets
cost = price * portfolio

Or some leet OLS. Oh yeah.

portfolio = (Q.t*Q).inv*Q.t*assets

Awesome eh? It probably seems more nifty to programmers and those who've tried matrices in Excel. Nasty. I even wrote a little Ruby library to make printing out the matrices prettier. Although probably if I count that extra time my average productivity drops quite a bit.

Ruby is pretty good all round. Kind of slow in spots. But that won't stop it from taking over the world pretty soon.

Why are governments the least risky people to give your money to when it comes to bonds, and the most risky people to give your money to when it comes to superannuation?

I'm enjoying finance, because I'm learning about a whole lot of stuff that I've never understood. But even so, a lot of it is really lame. We learnt about share price valuations yesterday. The way you work out how much a share should be worth is by putting into Excel how much all the future earnings of the company are going to be, and then pulling out of your arse how much you think the company should return in profits (like 5% a year or 15% if you feel like it). Then you divide something by something else and what you get a that is the "value" of the share. Pretty damn lame.