I was messing around with FRED data and noticed that the ratio of government expenditures to government receipts seems to show a dynamic equilibrium that matches up with the unemployment rate. Note this is government spending and income at all levels (federal + state + local). So I ran it through the model [1] and sure enough it works out:
Basically, the ratio of expenditures to receipts goes up during a recession (i.e. deficits increase at a faster rate) and down in the dynamic equilibrium outside of recessions (i.e. deficits increase at a slower rate or even fall). The dates of the shocks to this dynamic equilibrium match pretty closely with the dates for the shocks to unemployment (arrows).
This isn't saying anything ground-breaking: recessions lower receipts and increase use of social services (so expenditures over receipts will go up). It is interesting however that the (relative) rate of improvement towards budget balance is fairly constant from the 1960s to the present date ... independent of major fiscal policy changes. You might think that all the disparate changes in state and local spending is washing out the big federal spending changes, but in fact the federal component is the larger component so it is dominating the graph above. In fact, the data looks almost the same with the just the federal component (see result below). So we can strengthen the conclusion: the (relative) rate of improvement towards federal budget balance is fairly constant from the 1960s to the present date ... independent of major federal fiscal policy changes.
Footnotes:
[1] The underlying information equilibrium model is GE ⇄ GR (expenditures are in information equilibrium with receipts, except during shocks).
Movement toward federal budget balance is an improvement is it?
ReplyDeleteIf I am reading your charts right it rather looks like this "improvement" is always followed by a shocking increase in unemployment.
That is true, but it's similar to saying a spell of nice weather is always followed by a storm. But did the nice weather cause the storm?
DeleteI look forward to your teasing out some causality from your method. In fact, I've been looking forward to it for some time now. I'm beginning to wonder if it has any power to do so.
DeleteCausality goes from the shock to unemployment to the shock to the budget looking at the data above. There is an uncertainty in the time measurements (bigger for the G data), so incorporating that in maximum likelihood estimate tells us the positive unemployment (i.e. recession) shocks precede the positive G (i.e. deficit increasing) shocks.
DeleteYou can see that the arrows mostly precede the lines in the graphs.
Ah, it was not apparent that the exact position of the arrows is significant. So the arrows correspond to the shocks you have extracted from the unemployment data, whereas the lines are the budget shocks.
ReplyDeleteDo you have a post talking about shocks? In neoclassical economics they are the equivalent of saying 'God did it', and I am not religious.
From a basic understanding of MMT I would expect the causality to go both ways; rising unemployment forces up the deficit, a falling deficit increases unemployment (ceteris paribus). Can you spot this kind of dynamic?
Jason: “the (relative) rate of improvement towards budget balance is fairly constant ... independent of major fiscal policy changes”
ReplyDeleteI am open to any data which assesses the effectiveness of any policy. However, I don’t follow some of your argument here.
As you suggest, some of this is just basic logic. Governments have budgets, like anyone else, and their deficits will go up, compared with the counterfactual, when they face unexpected expenditure or have an unexpected shortfall of income (tax). If the budget assumes a certain level of unemployment, the budget will include a certain level of transfers to the unemployed. If unemployment goes up unexpectedly, the level of transfers will go over budget. Similarly, if the budget assumes a certain level of taxation from incomes, the rise in unemployment will probably reduce tax receipts to below budget as the unemployed no longer pay the tax they were expected to pay.
However, that’s where we part company. I have four main points.
First, one of the many crimes against humanity committed by macroeconomists is that they discuss policy using generalities like “monetary policy” and “fiscal policy”. Fiscal policy can mean many different things, so, we would have to look at each type of fiscal policy to draw any meaningful conclusions. For example, economists talk about “automatic stabilizers” as one type of fiscal policy. An example of automatic stabilizers is the automatic increase in unemployment transfers when unemployment goes up – which is what we see in the chart. The observation that deficits increase when unemployment increases is part evidence of the operation of fiscal policy!!
You could imagine an alternative system where a fixed amount was put aside in the budget for unemployment transfers and divided up amongst the number of unemployed. In that case, there would be no automatic stabilizers; no corresponding increase in the deficit when unemployment increased; and each unemployed person would receive a lower transfer when unemployment increased. Under that policy, the chart you are analysing would be different.
Second, the period on your chart starts around the period when Keynesian policies went out of fashion - apart from things like automatic stabilizers. For most of the period, economists advised policy-makers to focus on monetary policy, so I’m not clear what you mean by “independent of major fiscal policy changes”. Do you mean specific changes? If so, which ones and when did they occur? You may be correct here. However, you would need to be provide more detail before I could understand your argument.
(Continued)
Third, one of the points you make consistently, with which I strongly agree, is that you can’t just change a term in an accounting identity and use it to forecast a change in GDP. However, the same is true in reverse. You often can’t conclude, from a high-level chart showing a trend, whether that trend was due to a specific policy action or despite that action – unless you support your analysis with credible logic. That’s one of the reasons that this stuff is so hard.
ReplyDeleteTry a thought experiment. Suppose we see a chart of a human-designed system which shows the system has a constant velocity. Suppose we then change a setting on one of the system controls. Suppose that the velocity is unchanged after the control change. It feels to me that, based on this evidence alone, you would conclude that the control change had no impact on the system. That’s what you seem to be doing here regarding fiscal policy.
That is the wrong conclusion. The system in my thought experiment was a car and the control change was pressing the accelerator pedal. Now you will immediately see the problem with the earlier conclusion. The most plausible conclusion now is that the pedal was pressed as the car started up a hill. The difference between the two conclusions is that the second includes an understanding of the logic of the cause and effect relationships expected from using the accelerator.
Even though the system velocity did not change, it WOULD have changed had the pedal not been pressed. The same is often true when a government implements a policy. Maintaining the status quo is a success when the status quo is otherwise threatened. This often happens in business too. Apple and Samsung are constantly innovating the designs of their smart phones. This doesn’t necessarily lead to more sales for either. However, if one company stopped innovating, the other would take over its market share and reduce the viability of the original company, so they must both innovate just to stand still. The fates of Blackberry and Nokia are examples of what happens when companies do not keep up.
Remember that Paul Krugman argued for fiscal stimulus after 2008 as he thought that otherwise the economy would tank. The fact that the Great Recession was smaller than the Great Depression suggests to me that automatic stabilizers plus the extra stimulus had a positive impact. However, I’m not sure how you could prove this scientifically as there is no counterfactual against which to measure the difference.
Fourth, one of the things that is never discussed on the econ-blogosphere is how governments ascertain the effectiveness of fiscal policies when it is so difficult to assess effectiveness from ongoing high-level statistics. The answer is that they do controlled trials. The trials look for evidence for the effectiveness of the policy and evidence for the unintended consequences of the policy. For example, there are trials of universal basic income ongoing in Canada and Finland.
http://www.independent.co.uk/news/world/americas/ontario-canada-largest-province-launch-universal-basic-income-trial-kathleen-wynne-a7701451.html
http://www.independent.co.uk/news/world/europe/finland-universal-basic-income-trial-pilot-scheme-unemployed-stress-levels-reduced-a7724081.html
Governments do this because it is the most scientific way to assess policy effectiveness. These trials are analogous to the trials governments force drug companies to do before authorising a drug for sale to the public. Of course, governments continue to monitor policy effectiveness after the policy is fully implemented. However, they use detailed, specific surveys and statistics which are fit for purpose – not the high-level statistics in FRED. Again, this is analogous to drugs. No-one would attempt to analyse the effectiveness of an individual drug based on high-level summary health statistics.