Wednesday, 13 April 2016

Probability of Osborne Hitting his Surplus Target? Around 0%. Here's why.

Last month, when UK Chancellor George Osborne set out his budget, the Office for Budget Responsibility issued its outlook for the next few years. It judged that the probability of Osborne hitting his self-imposed target of achieving a budget surplus in 2020 was a little greater than 50%. Remember that Osborne not only wants to hit this target in 2020 but wants to make legislation that forces all governments to do the same in the future.

This is insane. The government must run deficits; I discuss here why we actually need to run much larger ones. And the probability of Osborne hitting that target without heavy financial engineering is, in my opinion 0% (rounded to the nearest percent).

Running a surplus is extremely rare. The reason is that for nominal GDP growth, more money must be spent every year than the year before. On top of this, the structure of our economy is such that we end up, on average, saving a small proportion of money from GDP each year, taking money out of the system. The saving needs to be replaced, and the extra demand for goods and services must come from somewhere.

On occasions this may come from high confidence leading to people spending more of their savings than the amount lost to new saving. But typically the money needed to grow the economy comes either from new private sector debt or new government debt.

The detail of my Demand Based Cash Flow model as well as my empirical findings are in my paper, which I recommend reading, but very briefly...  A 10% of GDP increase in private sector debt corresponds to around 1.5% economic growth in the year that it is taken out. Note that this multiplier is a lot lower than that commonly calculated for government debt (normally around 1), so government debt is much more stimulatory. And the bad news is that it exerts a drag on economic growth from then on. The drag is significant - maybe a 0.15% loss of  NGDP growth per year every year until it is repaid.

Private sector debt in UK, and worldwide, is now very high. This, in my opinion, is the reason for our current economic stagnation and the reason that we are at zero interest rates.

The mechanism that this works through is that the higher debt is, the more of GDP that goes to paying interest payments rather than wages. This is both by corporations and by individuals. On top of this higher debt means more demand for houses as well as lower interest rates - this leads to higher house prices and in turn higher rents.

This means even more money goes to savers/investors in an economy rather than workers. People tend to spend most of their wages but interest and dividends are often just accumulated in pension funds or other saving vehicles. So the marginal propensity to spend of workers is higher than that of savers.

For this reason, as debt has risen, demand has fallen. And the saving rate of the economy (which I define as the amount of GDP in year t that is not spent again in year t+1) has grown to around 3% of GDP each year.

Other factors contribute. The ability of corporations to achieve higher profit margins by suppressing wages is one.

And the intermediate step between both of these and lower NGDP growth is a shrinking of worker share of GDP (after rental payments).

I would like to thank Duncan Weldon for both the idea to look at this as well as the glorious spreadsheet that he directed me to with three centuries of economic data for the UK.

I made an estimate for worker share of GDP growth, following Weldon's example. The numerator is (compensation per employee x total people in employment) minus (estimated rent per property x total people in employment). The denominator is Nominal GDP minus (estimated rent per property x total people in employment). So an approximation for (wages minus rent), divided by (NGDP minus rent). The rent per property is estimated at 5% of the property value.

I plotted this worker share of GDP growth against the average government deficit for the subsequent ten years. Although in any one year the government deficit can be volatile, over a ten year period one can get an idea of how much demand needs to be replaced.

The graph looks like this:

It is not a perfect fit, but considering that it is just one factor in a complex system, it is reasonably close. I would argue that there does appear to be a strong relationship, especially if one ignores the immediate post-war period.

I also have private debt figures from 1963 onwards from BIS. The private debt level is -78% correlated to the above worker share measure. The level of private debt is also -61% correlated to the ten year average government balance. The three series appear to be linked; as predicted by my DBCF model. Also, much earlier, by Steve Keen, whose work is invaluable for more understanding of the role of debt.

From the graph it can be seen that with the current low level of worker share of GDP the economy is structurally saving considerably too much to allow the government to eliminate the deficit. Unless the worker share of GDP can be increased, and unless more money goes to people who will spend rather than save, then the economy can not bear the cost of a government surplus.

The graph actually suggests that deficits are too small. This, in my view, is the reason that NGDP growth is so much lower than trend. Higher deficits are really needed. The damage already done to the UK's productivity by Osborne's misguided policy is huge; this graph shows the sudden stop in 2008.
As Frances Coppola says, it is not the benefits to the disabled that are unaffordable but the chancellor himself. It could be argued that appropriate fiscal policy would have allowed a continuation of trend growth, in which case Osborne is responsible for a current annual loss of around 10% of UK GDP.

You may argue that a surplus has been achieved in the past so why can't it be done again? But there is a big difference here. When Gordon Brown in the UK or Bill Clinton in the US achieved budget surpluses they did so by allowing private sector debt to grow very quickly. At the time we had not hit the high levels of private sector debt and this was not too difficult. The budget surpluses were compensated for by lower interest rates and debt grew.

This is not possible now, though. Private sector debt has hit such a high level, and subsequently house prices are so high, that even zero interest rates can not stimulate enough new private sector debt to compensate for no government deficits. The current account deficit also makes it harder.

It is, of course, possible to get a surplus by simply cutting spending by enough. The problem for Osborne is that every cut makes the economy worse, leading to lower tax revenues. The target will just keep moving further and further away as he approaches it.

Now, I think that the increase in the minimum wage may make a small improvement here; only time will tell. But I also think that the benefits of this may be lost because of other budget measures that take from those with a high MPC (eg lowering tax credits) and give to those with low MPC (eg lowering inheritance and capital gains tax).

In conclusion, I think it totally impossible that Osborne reach his surplus target. I think that the Office of Budget Responsibility has models that do not properly include the feedback from lowering spending. And that what it calls budget responsibility is actually gross irresponsibility.


  1. Can't private bank debt substitute for a government surplus, at least as long as there's inflation?

    1. It can but it is usually a lot more inefficient, so much more is needed to get the same amount of growth. It also leads to asset price rises and instability in the economy.


Sorry, I have had to moderate comments because of an annoying pest spammer who keeps posting American football matches links.