Fuzzy logic did wonderful things in Japan. It was a term coined in the mid-1960s by the Azerbaijan-born mathematician Lotfi Zadeh to describe a new range-based logic that broke free of Aristotle's maxim that a thing either is or it isn't. Fuzzy logic allowed for varying degrees of truth. That enabled computer programmers to develop solutions for arrays of possibilities: a thing could be slightly true, mostly true, or even sometimes true.
Twenty years later, fuzzy logic sparked a revolution in Japan's consumer electronics industry. In the 1980s, Japan was overwhelmed by fuzzy appliances: TVs that adjusted their volume to the noise in a room, washing machines that detected how dirty your clothes were and rice cookers that compensated for sultry summer days. Fuzzy logic didn't work as well when the Japanese applied it to accounting and financial restructuring. When it turned out that much of Japan's property and investment bubble was predicated on the very slippery reasoning that stocks and property prices would rise forever and that rules on financial risk could be bent to accommodate new circumstances, it left the economy paralysed for a decade.
The global economy now suffers from a similar fuzz around the pricing of risk. Likewise, there are shades of Lotfi Zadeh haunting Dubai's efforts to pull out of its own financial quagmire. With so little in the way of hard, reliable data disclosed by authorities, investors, analysts and economists are left to draw fuzzy conclusions about whether things are looking mostly better or slightly awful. Facts are not enemies of confidence. Take the revelation by Dubai World last month that it had US$59 billion (Dh216.71bn) in liabilities. Many commentators saw this as a bombshell that raised questions about whether Dubai's estimated $85bn in debt might be a gross underestimate.
More careful examination revealed that Dubai World was reporting its consolidated liabilities - those not only at Dubai World, but the debts, bills and sundry other obligations at its roughly 21 subsidiaries and the more than 70 units they control. For its part, Dubai World owes $5.5bn, analysts say. Nakheel owes another $7bn, DP World $4.75bn, Dubai Drydocks $2.2bn, Jafza $2bn and Limitless $1.2bn. It's a lot of money. But it doesn't come close to $59bn.
The problem is that the only reason we know this is because of a level of sleuthing by analysts that would make a Kremlinologist proud. No one, outside the Government of Dubai, knows exactly how much Dubai and the companies it controls owe. "The lingering uncertainty over Dubai's creditworthiness stems at least in part from the fact that data on the emirate's external debt stock are not published by the local authorities," wrote Deutsche Bank economist Caroline Grady this month in a report that shed a rare beacon of clarity on the state of the nation's finances.
Lingering uncertainty explains why investors still regard Dubai as one of the six governments most likely to default. While improving global markets have helped lower the cost of borrowing for the UAE and for Dubai, Dubai's bonds remain one of the most expensive debts in the world to insure, ranking just above Iceland. By Deutsche Bank's estimates, Dubai will need to repay $6.8bn between now and the end of the year. Next year it will need to come up with another $10.1bn. It will need $12.1bn in 2011, $15.2bn in 2012 and $4.8bn in 2013.
How it will service that debt remains an open question. Some economists believe it will take years before trade, tourism and property, Dubai's lifeblood, recover to pre-crisis levels. Will Dubai need federal support until they do? If it does, can the Federal Government afford it? It turns out that most, if not all, the Dh50bn injected by the Ministry of Finance into the nation's banks back in February was borrowed from the Central Bank, analysts say.
This money came from the Central Bank at a time when its published assets were $48.3bn. The Central Bank then lent $10bn to Dubai by buying the emirate's bonds. Why lend Dubai US dollars? Most of Dubai's debts are in dollars, for starters. And many economists postulated that by doing so, the Central Bank might have thought that its foreign currency reserve levels would remain unchanged. They didn't. Lending dollars domestically turned $10bn in foreign-exchange reserves into domestic foreign-exchange assets.
But those assets didn't show up in the balance sheet for the quarter ending March 31 that the Central Bank submits to the IMF, according to Deutsche Bank. Not only did the Central Bank's foreign-exchange reserves not decline by the amount needed to buy Dubai's bonds, but the Central Bank's reported claims on official entities remained at zero. Where, Ms Grady asks, did Dubai's bonds disappear to and did it ever get the $10bn from the Central Bank?
More importantly, can the Central Bank afford to buy $10bn more? Dubai plans to sell a second $10bn before the end of the year. And given that Dubai has been unwilling to open its books to secure a credit rating, and that the Dubai Financial Support Fund it set up in July to administer the funds has said it won't say how they're used, few private investors are likely to want to buy them. That leaves the Central Bank as the only likely buyer. But with an estimated $24bn in forex reserves left, economists say buying another $10bn could leave the Central Bank perilously low on dollar reserves, particularly in a country that pegs its currency to the dollar.
This is where the fuzzy logicians leap in to cry "Oil!". Oil wealth will spring from the ground to fill the nation's wells of debt, they say. But nearly all of that wealth belongs to Abu Dhabi, and there is no explicit mechanism for transferring that wealth from Abu Dhabi to the Central Bank or to the Federal Government. We've all lived comfortably assuming that it can be, with no idea precisely how.
Until we have one, no amount of fuzzy assurances from officials can fully restore confidence. Solvency isn't a state of mind. Like pregnancy, a government cannot be mostly solvent, almost solvent or sometimes solvent. It either is or it isn't. firstname.lastname@example.org