ECB researchers find fiscal policy is very effective and more so if central banks buy up the debt

The ECB published a Working Paper recently (September 2021) – Monetary and fiscal complementarity in the Covid-19 pandemic – which represents progress in the narrative. While the technical model that the ECB uses is just an ad hoc attempt to reverse engineer the reality so they can claim they can explain it, what is useful from the exercise is that the old mainstream narratives that fiscal policy is ineffective in providing permanent boosts to real output (or that austerity does not permanently damage the growth trajectory) can no longer be sustained. The taboo surrounding central bank purchases of government debt because they cause accelerating inflation can no longer be sustained. The claims that fiscal deficits drive up interest rates can no longer be sustained. Now the public debate just has to reflect that reality and we will have made progress. Of course, this is all core MMT – we knew it all along!

The ECB authors, Jagjit S. Chadha, Luisa Corrado, Jack Meaning, and Tobias Schuler analyse the implications of the way that monetary policy is interacting with expansionary fiscal policy.

They note that in the US we see “the Federal Reserve System purchasing extraordinary quantities of securities and the government running a deficit of some 17% of projected GDP”.

And the US central bank has also “pushed the discount rate close to zero” and has provided banks “with emergency liquidity … through a new open-ended long-term asset purchase programme.”

Their overall conclusion that is this approach:

… the central bank uses reserves to buy much of the huge issuance of government bonds and this offsets the impact of shutdowns and lockdowns in the real economy. We show that these actions reduced lending costs and amplified the impact of supportive fiscal policies.

And if monetary and fiscal policy was not working together for once, the US would have “experiences a significantly deeper contraction as a result from the Covid-19 pandemic”.

The point is obvious and runs counter to the way in which mainstream macroeconomics has been taught and practiced in this New Keynesian era, where monetary policy was assigned the primary role (adjusting interest rates) and fiscal policy was deemed to be passive and biased towards surplus creation.

It also establishes that fiscal policy is highly effective in reducing the negative consequences of a significant non-government spending reduction.

Which runs counter to the received wisdom of the standard New Keynesian approach.

The NK approach to fiscal policy is exemplified by the 1993 textbook of John Taylor – see Macroeconomic Policy in a World Economy: From Econometric Design to Practical Operation (Norton) – or the analytical model presented by Frank Smets and Rafael Wouters (2007) – Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach (American Economic Review, 97 (3), 586-606).

The Taylor approach characterises the response of the economy to a fiscal stimulus as being immediately positive (typically) then diminishing more or less quickly as the crowding out impacts of higher interest rates bite on private sector investment and acceleration inflation squeezes real profits.

In the Smets-Wouters approach the crowding out is more or less immediate as the government injects each new dollar into the economy.

Private consumption and investment expenditure immediately declines and mostly offsets the fiscal stimulus.

The point is that within the NK paradigm there are less and more extreme versions in the short-run after a fiscal stimulus but they all converge on the same result in the long-run – that fiscal policy essentially has no long-run impact on output but worsens the inflation profile of a nation.

In relation to yesterday’s blog post – They never wrote about it, talked about it, and, did quite the opposite – yet they knew it all along! (September 20, 2021) – the mainstream economists who claim that Modern Monetary Theory (MMT) is unnecessary because the standard NK framework is perfectly capable of allowing a fiscal stimulus in the short-run to deliver positive output effects, will all converge on this long-run ineffectiveness conclusion.

The debate within NK economics is the speed to which the economy converges on the long-run equilibrium after a shock and how strong the short-term impact of a stimulus turns out to be.

But make no mistake, the conventional paradigm in economics believes that fiscal policy stimulus has no long-term positive impacts and only, ultimately invokes a higher inflationary path.

That is what the ‘we knew it all along’ crowd teach and build into their research models.

Further, the NK economists are not really macroeconomists at all, despite them claiming to be.

Why would I say that?

I discussed this point in this blog post – Mainstream macroeconomic fads – just a waste of time (September 18, 2009).

The reason is that they build their aggregate framework (what they call the ‘macroeconomic’ level) from simple maximising, microeconomic principles beginning with an individual.

That micro level of analysis yields certain conclusions – for example, a single firm might benefit if its workers took a pay cut because while the unit costs the firm would face would fall, it would be unlikely that the damage to sales would be significant if no other workers took a similar pay cut.

Whether that increased employment for the firm or whether it just pocketed the increased gap between costs and revenue is moot.

But if all firms tried the same strategy – cutting wages – unit costs would fall across the economy but so would incomes and consumption expenditure which would damage sales, and, probably push the economy into recession.

This is the famous case of the fallacy of composition, which I analysed in this blog post (among others) – Fiscal austerity – the newest fallacy of composition (July 6, 2010).

Consequently, treating macro as if it is micro means there is a tendency to conclude that what applies at the individual level also applies at the aggregate level, which is demonstrably false.

Why is this important?

It is important because the NK approach to ‘macroeconomics’ claims the higher authority because it says it is derived from consistent, microeconomic optimising principles – a sort of appeal to technical superiority, which they claim the old-style Keynes approach lacked.

The problem then is that to escalate the single consumer/firm micro analytical results up to economy-wide relationships – all households, all firms, all industries etc – the link between the micro optimisation and the aggregate proves to be impossible to achieve.

To resolve that problem, the NK approach creates the ultimate fudge – it assumes what they call the ‘representative agent’ – which they assert obeys the same behaviour and motivations as the micro optimising agent.

So there is one ‘infinitely-lived household’ or one representative firm used in the framework at the macro level.

Which means that, in fact, they never really leave the world of neoclassical microeconomics and in trying to assert any results about what happens at the aggregate level they fall into the fallacy of composition trap.

The ‘we knew it all along’ crowd don’t often admit to that do they? They just pretend to be doing macroeconomics.

If we dig deeper, we find that NK models essentially predict financial instability and accelerating inflation will inevitably accompany a zero interest rate policy deployed by a central bank.

The NK approach is an amalgam of what has been referred to as fixed wage Keynesian economics and the classically-inspired real business cycle theory based on rational expectations.

I won’t go into the details of that conflation here except to say that it effectively abandons everything that is Keynes from the Keynesian part and replaces it with the old Classical beliefs that governments cannot change the course of the real economy in the long-run and only influence nominal variables (inflation etc) if they try.

The rational expectations influence really began with the 1975 publication by Thomas Sargent and Neil Wallace – Rational Expectations, the Optimal Monetary Instrument, and optimal Money Supply Rule (Journal of Political Economy, 83, 241-254).

When I started studying economics in the mid 1970s, this was a raved about paper. I read it and couldn’t believe how asinine it was. But that was the times.

Sargent and Wallace effectively established the framework that permeates NK economics to this very day.

If we took the model seriously then Japan would have hyperinflated two or more decades ago

The ECB find that fiscal policy is very effective and more so when the central bank ‘funds’ it

Anyway, the ECB paper shows that the world is moving beyond this moribund framework, which is a good sign.

They seek to conjecture about the following juxtapositions:

1. A large supply shock (the shutdown).

2. A large decline in money velocity – how much the money supply turns over in transactions per period (the lockdowns stifling expenditure).

3. A massive fiscal stimulus – accompanying by debt-issuance.

4. The stimulus accompanied by the central bank buying the debt issued with credits to bank reserve accounts.

The hard-core NK framework predicts rising interest rates, accelerating inflation, and, only short-term real output gains followed by falls in household consumption expenditure (rising saving to pay for the implied higher taxes to pay back the public debt increase) and business investment (as the rising interest rates crowd out non-government borrowing), which, ultimately, undermine any temporary output gains.

The reality that the ECB authors want to try to understand is why the NK predictions systematically failed.

They find that, in fact:

… the provision of reserves stabilised the value of collateral and amplified the impact of supportive fiscal policies … the fall in output in the first stage of the pandemic might have been as much as twice as large, with a significant deflation, loss of employment and falls in asset prices, if such extensive fiscal and monetary policies had not been implemented.

Which is the standard result that core MMT has come to 25 years ago!

I might be tempted to say ‘we knew it all along’ but I won’t (-:

The US situation is thus:

1. Total US public debt has risen by $US5,328.1 billion since the December-quarter 2019.

2. The US Federal Reserve Bank holdings of the debt since then has risen by $US3,007 billion.

3. Which means that the US Federal Reserve system of banks has purchased around 56.4 per cent of the debt issued over the pandemic.

4. The US Federal Reserve holdings has risen from 11.4 per cent to 19.8 per cent (by June-quarter) of all outstanding public debt.

That is quite a shift.

The following graph shows the proportion of outstanding US public debt held by the US Federal Reserve Banks since the March-quarter 1970.

You can see that over the Monetarist period, the proportion fell steadily and then the two big jumps coincided with the GFC and the pandemic.

Long-term US bond yields remain low.

Have a look at the history of the US 10-year Treasury bond yields since the beginning of 2021.

You can get for all available maturities from the US Department of Treasury’s site – Daily Treasury Yield Curve Rates.

As the economy started to opened up a bit in February and sentiment improved, investors started to diversify their portfolios away from the risk-free Treasury bonds and yields rose a little.

This set the mainstreamers off into a conniptive-fever (don’t look that word up as I just made it up. Etymology – derived from conniption or hysterics).

They started to increase their attacks on MMT economists like me with the ‘I told you so’ banter.

Well then mid-March came along and yields fell again and they have been largely flat since the middle of July.

Nothing going on here is the message.

Another way of looking at this is shown in the next graph, which shows the US Treasury yield curve across all maturities since the start of the 2021 – at various snapshots.

The longer end certainly rose in the first six months but since then, the yield curve has flattened rather than steepened.

In this blog post – Rising prices equal an inflation outbreak (apparently) but then the prices start falling again (June 21, 2021) – I explained why these trends militate against the accelerating inflation narrative.

If investors expect that inflation is becoming an issue, then they will demand higher yields at the primary issue and will be prepared to pay less for outstanding bonds in the secondary market.

The higher the expected inflation, the higher the risk premium that will be built into required yields.

The facts thus do not support the mainstream ‘inflation’ narrative.

The ECB authors tried to come to terms with these results that run counter to the standard NK framework – that used by the ‘we knew it along’ gang.

The combination of a declining GDP (total sales in the economy) and the expansion of bank reserves and bank deposits (coming from the fiscal stimulus) has reduced the velocity of money in the US (the turnover rate of the stock of broad money).

The ECB authors note that this was due to households cutting back “spending sharply” and increasing their saving.

They write:

Total household income did not fall by nearly as much as spending, largely because those who are still employed, working from home or elsewhere cut back on their purchases. Savings jumped as a result …

Fiscal policy thus had significant increased space in terms of real resources to operate in.

Remember, MMT defines fiscal space in terms of idle real productive resources rather than in terms of current numbers pertaining to deficits/surpluses or public debt.

Fiscal space is not a financial concept but a real resource concept, which is totally at odds with the way the IMF and conventional economics defines it.

And the US Federal Reserve bought up a significant quantity of the new debt issued, which meant the private investors really no longer determined yields.

The technical model that the ECB has contrived is designed to reverse engineer the empirical reality.

I wouldn’t study it in detail.

The point is that they have had to fundamentally alter the standard NK approach to generate the intended results.

But the conclusion is inescapable.

1. “A combined fiscal-monetary response may have helped avoid turning the Covid-19 crisis into an economic recession of even greater magnitude and severity”.

2. “if the Federal Reserve had not intervened, output would have fallen by more than 10% more on impact and in the following quarter”.

3. “Real wages would be down by more than 15% more and unemployment up by more than 20%. Wages would be 20% lower than with QE. As a result inflation would have fallen even further.”

4. “we find that prompt, combined fiscal-monetary interventions mitigated the impact of the pandemic shocks and helped to establish a more rapid recovery to pre-crisis levels of activity.”

Conclusion

The point is that the old mainstream narratives that fiscal policy is ineffective in providing permanent boosts to real output (or that austerity does not permanently damage the growth trajectory) can no longer be sustained.

The taboo surrounding central bank purchases of government debt because they cause accelerating inflation can no longer be sustained.

The claims that fiscal deficits drive up interest rates can no longer be sustained.

Now the public debate just has to reflect that reality and we will have made progress.

Of course, this is all core MMT – we knew it all along!

That is enough for today!

(c) Copyright 2021 William Mitchell. All Rights Reserved.