Never reason from an accounting identity: Why Thomas Piketty is wrong about Osborne's plan

This is a guest post written by Simon Wan.

Last week, George Osborne announced that he would entrench in law the government's commitment to run long-term budget surpluses. This is, admittedly, an ill-advised policy. But the response from Osborne's critics was even worse. In a letter published in the Guardian, a roll call of the government's main detractors on economic policy argued against the move. And in the process they broke one of the fundamental rules of social science––never reason from an accounting identity. 

For me, the key sentences from the Guardian letter are these:

The government’s budget position is not independent of the rest of the economy, and if it chooses to try to inflexibly run surpluses, and therefore no longer borrow, the knock-on effect to the rest of the economy will be significant. Households, consumers and businesses may have to borrow more overall, and the risk of a personal debt crisis to rival 2008 could be very real indeed.

A little bit of background here: by definition the net financial asset (or debt) position of the total economy is zero because any debt owed by one person is someone else's asset. So, according to this line of reasoning, if the government reduces its net debt, then the rest of the economy (the private sector and, in particular, households) must be reducing its net assets, in the process becoming poorer and possibly even having to borrow more.

First, there are some easy shots we can take at this argument. It ignores the existence of a foreign sector, which, in the UK's case, is not insignificant. Around 30% of outstanding gilts are held by overseas residents, meaning that any net saving by the public sector will at least partly be born by dissaving abroad. A point often made is that the last financial crisis was caused by global imbalances driven by persistent current account deficits in developed markets and savings gluts in emerging economies. Wouldn't lower borrowing in the West go some way towards remedying this? Furthermore, many countries run long-term budget surpluses without impoverishing the domestic economy. Norway, for instance, has one of the largest sovereign wealth funds in the world. The reason why this does not result in a large debt pile for Norwegians is of course that the country's oil wealth is invested almost entirely overseas.

More importantly, accounting identities are truisms, making arguments based on them largely meaningless. The claim that reducing government borrowing will reduce private sector wealth is as spurious as saying that the government should borrow as much as possible so as to boost private sector wealth. What's gone wrong here of course is that we are not just talking about private sector wealth (or debt), we are talking specifically about the private sector's wealth with respect to the government. So, yes, reducing the state's debt will decrease the private sector's claims on the government, but there is no reason why this means households would have to borrow more. Certainly there is no reason why it would cause households to borrow more from each other or from private lenders. And of course, it is patently untrue that when governments pay back debt they are reducing the wealth of the private sector. Indeed, this just changes the composition of assets held by the private sector from bonds and bills into broad money.

Taken to its logical conclusion, the argument presented in the Guardian letter would call for the government to privatise all its financial assets (since these are claims of the state on the private sector, and therefore selling them would raise the net wealth of the private sector with respect to the government). I don't see Messieurs Chang, Piketty or Murphy enthusiastically calling for the British government to sell its remaining stake in Royal Mail.