As the danger of a global depression recedes, the themes I am picking up regularly now from commentators, politicians etc are all pointing back to the mainstream status quo version of the way the economy works, in particular, for the purposes of this blog the labour market. I expect to increasingly hear and read the rhetoric that dominated the public debates prior to the crisis – that unemployment is essentially a supply-side phenomenon reflecting choices made by individuals in the context of government welfare policy that distorts these choices in favour of not working. In this context, the simple act of extending unemployment benefits in the US has been controversial. This takes us back to the dominant debates over the last 20 years which saw governments all around the World pursuing policies that were antithetical to full employment and pernicious in their impact on the victims of their policy failures. Stay tuned – 2011 – the mainstream will be in full attack mode again – conveniently forgetting where we have been over the last 3 or so years.
In recent days, there has been a kerfuffle about statements made by Republican senator for Arizona John Kyl last week (March 1, 2010) which can be read from the Congressional Record for the Senate debate about the request to extend unemployment benefits in the US.
I have to say that in one sense I agree with Kyl rather than the so-called progressive response to him. Which makes for an interesting tension I would suggest. More about which later.
Here is some of what he said (you can read the full account by going to the link I provided above to the Congressional Record):
I wish to address the bill that is on the floor. The bill has been denominated by my colleagues on the Democratic side as a jobs bill, but it will not create any new jobs and when considered in conjunction with the health care legislation the President has proposed will actually cost jobs and I wish to address that …
What of the subject of unemployment coverage extension which we have just been debating? That doesn’t create new jobs. In fact, if anything, continuing to pay people unemployment compensation is a disincentive for them to seek new work. I am sure most of them would like work and probably have tried to seek it, but you can’t argue it is a job enhancer. If anything, as I said, it is a disincentive …
So it is not a jobs bill, and it is beyond me how it could be denominated as such.
He was then attacked by a Senator from Montana who said that with the Unemployment to vacancy ratio above 5 it was “ridiculuous” to claim that “unemployment insurance is a disincentive to jobs”.
Kyl then responded:
… I said it is not a job creator. If anything, it could be argued it is a disincentive for work because people are being paid even though they are not working. I certainly did not say, and would never imply, that the reason people don’t have jobs is because they are not looking for them. It is true that a lot of Americans have gotten so tired of looking for jobs or believe they are not going to find them that they have stopped looking and, as a result, the unemployment numbers are probably higher than the roughly 10 percent that is quoted now. Some people believe it could be as much as 17 percent. This is why I have supported every extension of unemployment benefits. I have voted for them. As my colleague says, there are five people looking for every job that exists. If they cannot get the jobs, they needed support.
But what I said is true, and if my colleague can find a source that says it is not true, show me. But providing unemployment benefits doesn’t create jobs. The bill we have before us is denominated around here as a jobs bill. That is the biggest single expenditure in the bill, and it doesn’t create jobs.
In between these exchanges there were other statements from Kyl.
He spoke at length about what he considered to be Democratic chicanery that they were trying to exempt the bill from the “pay-go” rules which require that if the US Government intends to spend money they have to “find an offset in the form of a spending deduction or revenue enhancement that covers the cost of that new spending.”
All Kyl demonstrated in that exchange is that he doesn’t understand that the US government faces no intrinsic financial barrier in being able to pay the unemployment benefits without any offset and that the pay-go legislation is always likely to be counter-productive in the sense that such fiscal rules do not allow governments to react fully to a crisis that requires counter-stabilising net spending.
He also claimed that because Obama had said they would have to put up taxes to pay for the new spending that it would reduce jobs. All revealing a failure to understand macroeconomics.
On March 4, 2010, Paul Krugman attacked Kyl in his column Senator Bunning’s Universe.
For non-US readers who haven’t been following this debate – Bunning is Republican Senator Jim Bunning of Kentucky – who stalled the debate on the extension bill and generally showed contempt for the unemployed throughout the proceedings.
Krugman claims that the debate showed the:
… incredible gap that has opened up between the parties. Today, Democrats and Republicans live in different universes, both intellectually and morally … What Democrats believe is what textbook economics says: that when the economy is deeply depressed, extending unemployment benefits not only helps those in need, it also reduces unemployment. That’s because the economy’s problem right now is lack of sufficient demand, and cash-strapped unemployed workers are likely to spend their benefits. In fact, the Congressional Budget Office says that aid to the unemployed is one of the most effective forms of economic stimulus, as measured by jobs created per dollar of outlay …
In Mr. Kyl’s view, then, what we really need to worry about right now – with more than five unemployed workers for every job opening, and long-term unemployment at its highest level since the Great Depression – is whether we’re reducing the incentive of the unemployed to find jobs. To me, that’s a bizarre point of view – but then, I don’t live in Mr. Kyl’s universe.
First, Kyl supported the bill and in his statement above he clearly indicates that he thinks that if the unemployed “cannot get the jobs, they needed support”. So I think Krugman is fitting the wrong criminal.
Second, Kyl is both correct and incorrect about the extension of unemployment benefits not creating any jobs. Where he is correct is that the measure proposed by the Democrats is definitely not a “Jobs Bill” (to quote Kyl).
To see this, you should understand that he is wrong (as per Krugman’s intervention) because he doesn’t understand automatic stabilisers. The provision of income support works automatically to stop aggregate demand falling as far as it would in their absence. It is the opposite of the tax system – revenue collapses which increases the budget deficit which is expansionary in itself.
But to claim that income support that ekes out via the automatic stabilisers is a strategy for job creation – which is what the Democrats have been suggesting and which is implied by Krugman – is to deprecate the concept of a job creation program beyond recognition.
Third, this bears on Krugman’s claim that the Republicans are heartless while the Democrats are moral types. The fact that the US Government – currently controlled by the Democrats has allowed unemployment to rise to 10 per cent (17 per cent if you include those who have given up looking) and are only responding by extending the pitiful income support available to the victims of the aggregate demand collapse tells me that there isn’t all that much difference between the two parties.
Both parties’ actions are being driven by a fundamentalism about deficits and debt – which reflects the common misunderstanding about the way the modern monetary system operates and the opportunities the fiat currency system provides to a currency-issuing government, such as in the US.
It is this basic error in reasoning that is common across both parties and really dominates the more superficial differences that one might observe in demeanour and policies.
Not to put too finer point on it – statements from the Obama Administration over the last year – “the US has run out of money” etc – have exemplified the best rhetoric that the deficit terrorists can muster.
To see how restricted the US government’s stimulus response has been you might like to read this article in yesterday’s (March 7, 2010) Financial Times – Good for America, as far as it went by Clive Crook who says that “Three-quarters of the electorate thinks the stimulus was mismanaged. The country’s children are being strung with debt, voters reckon, for no good reason.”
Crook however concludes that:
The public is wrong about the stimulus, but the error is understandable. As with healthcare reform, the machinations that produced it were gruesome. As with healthcare reform, the administration had no clear olicy of its own, and relied on a dysfunctional Congress that the country does not trust.
Two other factors intervened. First, the downturn was deeper and more tenacious than expected. Second, not so well understood, the stimulus was smaller than it looked.
It is clear that the early responses were inadequate and reflected the neo-liberal ideology running throughout the Obama Administration. The fact they thought monetary policy would be effective as a first reaction demonstrates their failure to appreciate both what was happening and what tools were appropriate.
As an aside, Australia’s fiscal response was early and significantly larger than the US government’s response and that helps explain why we largely avoided the major meltdown that other nations are still struggling through.
Crook summarises the Congressional Budget Office data which shows that:
… the stimulus boosted output in the fourth quarter of 2009 by between 1.5 per cent and 3.5 per cent, and reduced the unemployment rate by between 0.5 percentage points and 1.1 percentage points – unspectacular gains, given the scale of the commitment, but valuable nonetheless.
Crook also cans Barro’s claims which I addressed in this blog – Pushing the fantasy barrow.
But his most interesting observation (quoting an NBER study) which helps explain the extreme lassitude that the US economy is wallowing in at present was that:
… the aggregate fiscal expenditure stimulus in the US, properly adjusted for the declining fiscal expenditure of the 50 states, was close to zero in 2009.
Would the situation have been worse under the Republicans? I doubt it. So the claim that the Democrats are the nice ones who care about the unemployed and are actively seeking stimulus measures turns out to be false – at least in 2009.
However, Kyl is also misguided about the role of unemployment benefits. Here is some ways of thinking about that question.
UV Curve for the US
Economists have long used the unemployment-vacancy (UV) relationship, the so-called Beveridge curve, which plots the unemployment rate on the horizontal axis and the vacancy rate on the vertical axis to investigate these sorts of questions.
The logic is that movements along the curve are cyclical events and shifts in the curve are alleged to be structural events. So for many countries the UV relationship has shifted several times over the last 40 years or so.
However, the notion that there is a neat decomposition between shifts in and movements along the curve is highly contested and has not been reliably established in the empirical or theoretical literature.
One of my earliest papers, which came from my PhD work was published in 1987 – ‘The NAIRU, Structural Imbalance and the Macroequilibrium Unemployment Rate’, Australian Economic Papers, 26(48), pages 101-118 – showed that structural imbalances (supply constraints) can be the result of cyclical variations and can be resolved, in part, by attenuating the amplitude of the downturns using fiscal policy.
In other words, there is no decomposition as the mainstream would like us believe.
This is reinforced by work I did a decade ago where it is clearly shown that where the U-V curve shifts, almost always the shifts were associated with major recessions which generated structural-like changes in the labour market. In other words, the shifts are driven by cyclical downturns (aggregate demand failures) rather than any changes in autonomous supply side behaviour (like worker attitudes changing, or welfare policy introducing distortions to incentives, etc).
However, the mainstream interpretation which Senator Kyl is just rehearsing is that the shifts are due to a failure of the unemployed to seek work as effectively as before.
This is the classic line taken by Layard, Nickell and Jackman in their 1991 book which was extremely influential at the time and basically provided the theoretical case for the OECD Jobs Study (1994) which served as the blueprint for labour market deregulation since the early 1990s and underpinned the pernicious welfare-to-work policies that many governments introduced in the last 15 years.
Layard and Co claimed that outward in the European Beveridge curve, for example, represented a “a fall in the search effectiveness among the unemployed”. They repeatedly have claimed that the persistent unemployment over the last 20 years is the result of workers becoming “too choosy”.
Since that time, the OECD has been constantly pressuring governments to abandon the hard-won labour protections which provide job security and fair pay and working conditions for citizens. They also deemed welfare payments including unemployment benefits to be disincentives to work.
Please read my blog – The OECD is at it again! – for more discussion on this point.
An examination of the empirical evidence over the period would suggest that the major claims made by Layard and Co (and the myriad of hangers-on that have followed them) do not bear scrutiny.
While they claimed declining search effectiveness distorted by excessive generous welfare payments caused the rising unemployment rates prior to this recession, it is highly probable that both are caused by insufficient demand for labour. The policy response then is – of-course – entirely different and would emphasis fiscal policy expansion underpinning job creation.
But this argument cannot even be made in the current downturn. The following graph shows the US Beveridge curve since December 2000. The vacancy rate is from the US Bureau of Labour Statistics JOLTS data for job offerings which only begins in December 2000. The unemployment rate is also taken from official BLS data.
Even the mainstream “text book” case would tell you that the data is recording a cyclical event with the sharp movement down a static U-V relationship – this sort of movement indicates a fall in job vacancies and a corresponding rise in unemployment – both being the manifestations of a collapse in aggregate demand and a failure of government policy interventions to redress it.
To examine this further, I had a look at some phase diagrams.
I first started using phase analysis in this 2001 paper – Exploring labour market shocks in Australia, Japan and the USA, which has since been published but you can get the free working paper version at the link provided. You will find reference in that paper to other relevant material if you are interested.
The following two phase diagrams show the current values of the respective time series plotted on the y-axis against the lagged value of the same series on the x-axis. The annual data is from 1960 to 2009 (so the graphs are for 1961 to 2009 given the lagged values).
The first graph shows the phase diagram for the US unemployment rate while the following graph shows the phase diagram for vacancies.
The vacancy data is derived from the Help wanted advertising index (2005=100), compiled by The Conference Board and published as part of the OECD Main Economic Indicators. I could have used the US Bureau of Labour Statistics JOLTS data for job offerings to generate a similar diagram but as noted above the dataset only begins in December 2000 and so doesn’t show you the past cycles.
These scatter plots are helpful in four distinct ways. First, the charts provide information on whether cycles are present in the data.
Second, the presence of “attractor points” can be determined. The points might loosely be construed as the middle of each of the traced-out ellipses.
Third, the magnitude of the cycles can be inferred by the size of the cyclical ellipses around the attractor points.
Fourth, the persistence (strength) of the attractor point can be determined by examining the extent to which it disciplines the cyclical observations following a shock. Weak attractors will not dominate a shock and the relationship will shift until a new attractor point exerts itself.
So if you examine the unemployment phase diagram first, you can see that US labour market has fluctuated around an attractor unemployment rate of around 5 to 5.5 per cent although the magnitude of the cycles around it has been variable. The early 1990s recession, while significant, did not promote a new attractor nor has the current recession.
This is quite different to say Australia where the attractor shifted in the 1974-76 period outwards (during the major recession), and the two subsequent recessions (1982 and 1991) oscillated around the higher point with varying cyclical magnitude. The recent downturn has not caused a shift in the current attractor in Australia – probably because it was a relatively mild recession.
It is clear in the US case that the economy takes several years to recover from a large negative shock even if the attractor remains constant.
But it is clear that the direction of the dynamics during a recession is to generate loops that head in the North-East direction then resolve slowly via a south-west adjustment.
While the vacancy rate attractor for the US has not exhibited any notable shifts over the period examined (since 1960) the cycles have been of different magnitudes.
The current downturn however suggests that a new attractor might be forming such is the collapse in the demand-side of the US labour market. So if anything there is a movement downwards in the relationship.
The mainstream text-book supply-side analysis would interpret any upward movement the unemployment relationship as a decline in labour market efficiency.
But equally, they would interpret the downward shift (or tendency) in vacancy relationship as increasing matching efficiency. Clearly, both states cannot hold.
The only consistent interpretation for the dynamics shown is provided by the view that the US economy is now severely demand constrained as a result of the collapse in private spending and the failure of the government response to adequately stimulate demand.
Any endogenous supply effects that may have occurred in skill atrophy and work attitudes – for example, workers giving up looking for work and instead trying to tide themselves over with the receipt of unemployment benefits – are not causal but reactive.
There has been a huge number of research articles published in the self-serving mainstream journals supposedly proving that unemployment support reduces employment by undermining the incentives of the jobless to search for work.
If you read this literature you will realise it proves nothing of the sort. The literature is replete with fudged datasets, spurious techniques and other niceties that the mainstream researchers use to cheat on their findings or mislead the reader as to what is actually being demonstrated.
The most sustainable argument that can be maintained however is quite different. It is summarised by this nice quite from Michael Piore (1979: 10):
Presumably, there is an irreducible residual level of unemployment composed of people who don’t want to work, who are moving between jobs, or who are unqualified. If there is in fact some such residual level of unemployment, it is not one we have encountered in the United States. Never in the post war period has the government been unsuccessful when it has made a sustained effort to reduce unemployment. (emphasis in original) [Unemployment and Inflation, Institutionalist and Structuralist Views, M.E. Sharpe, Inc., White Plains]
Despite all the labour market and related supply-side reforms that have been introduced across the OECD countries over the last 15 or so years, the unemployment rate persists at high levels due to demand deficiency.
The conventional NAIRU approach tends to neglect the role of aggregate demand and focuses on the supply side. The empirical foundation of the NAIRU is without credibility – Please read my blog – The dreaded NAIRU is still about! – for more discussion on this point.
And if the government isn’t willing to provide jobs for the unemployed, which is the best option when private demand collapses, then the only moral thing to do is provide indefinite income support at a level sufficient for the unemployment worker to avoid falling into poverty.