Green QE and the Juncker Plan: a response to Richard Murphy

Richard Murphy proposes what he calls "green quantitative easing" to support the Scottish government's plans for fiscal expansion.

I've criticised the "green QE" proposal before. But this is a particular framing of it that raises some interesting issues about the nature of currency unions and the purpose of monetary policy.

Here's the heart of Richard's proposal:
It is worth explaining what I think the SNP is referring to when it mentions ‘innovative finance mechanisms’. It is my belief that they are referring to green infrastructure quantitative easing,which was the subject of a speech I made at the Convention of Scottish Local Authorities conference in March. In that speech I made clear that if the SNP wanted to do a service to Scotland, and to the rest of the UK, it would use its bargaining power after May 7 to demand that the UK government create a new form of quantitative easing that would be quite deliberately intended to provide the funding for new investment in infrastructure, housing, new energy systems, transport and the other essential underpinnings of growth in Scotland and throughout the UK.
This form of quantitative easing is entirely possible. As an example, the proposed Scottish Development Bank could, using this form of quantitative easing, issue bonds to the finance markets (because that is what EU law requires) that could then be repurchased by the Bank of England using funds specifically created by for this purpose.
So, Richard's proposal has two parts:
  • bond issuance by a government agency to fund infrastructure projects
  • purchase of those bonds by the central bank using newly-created sovereign money.
Part 1: Bond issuance

The first part is eminently sensible. The UK has a desperate need for infrastructure development, while institutional investors such as insurance companies and pension funds have a desperate need for long-term investments giving low stable returns. There is clearly an opportunity for the next UK government to seize.

Furthermore, the crying need for regional rebalancing in the UK suggests that a network of regional development banks might be a better approach to the investment crisis than a single UK-wide institution. So it makes sense for the next UK government to create a network of regional Development Banks, of which the proposed Scottish Development Bank would be one. These regional Development Banks would issue their own bonds to fund infrastructure and other long-term developments in their regions.

So there is in principle no problem with the idea of a Scottish Development Bank issuing bonds to finance infrastructure, housing, renewable energy systems and transport improvements in Scotland, as Richard suggests. The framework would be similar to the EU's Juncker Plan: the Scottish government would provide capital to the Scottish Development Bank, which would then leverage that capital by issuing bonds to the private sector.

However, I would at this point issue a strong warning to a Scottish government interested in "innovative financing mechanisms". Innovative finance has a way of turning round and biting you if you don't pay attention to where the risk ends up. Juncker's plan is based entirely on public sector guarantees and leveraging of funds already committed to other enterprises. This is a recipe for disaster. State-owned banks can fail if they don't have enough real capital. And it isn't safe to assume that there will be no significant losses on long-term infrastructure projects. My uncle owned subordinated bonds issued by Eurotunnel in its very early days, some of which I inherited on his death. After years of losses they were eventually written down to zero. So a Scottish Development Bank must have sufficient REAL capital - i.e. Scottish taxpayers' money uncommitted elsewhere - to provide a reasonable cushion for leveraged infrastructure finance.

Part 2: Monetising the bonds

This is where the second part of Richard's proposal comes in. He envisages the creation of what we could describe as a "pump". The Development Bank would issue the bonds: the Bank of England would buy them: interest and repayments would be repatriated to the Government and effectively written off. It would, in short, be completely circular. The debt would be off the Government's books, any losses would be absorbed by the Bank of England and the new money would reflate the economy. What could possibly go wrong?

In a follow-up to my post about Juncker's synthetic CDO, I described such a "pump" mechanism involving the EIB and the ECB. And I suggested that the Juncker Plan had been concocted by Juncker and Draghi jointly with the intention of taking advantage of the ECB's forthcoming QE programme. There was no doubt at the time that the ECB planned to do QE, and EIB bonds are eminently suitable for ECB purchase because of their triple-A rating. Increasing EIB issuance to fund much-needed Eurozone infrastructure investment would be an effective way of channelling QE money directly to where it is needed. It's a very clever scheme.

But Richard's plan, although apparently similar to Draghi's Machiavellian scheme, is in fact fundamentally different. And it carries huge risks.

Let's suppose that the SNP does succeed in persuading the Westminster government to instruct the Bank of England to buy infrastructure bonds. To keep it simple, I'm going to assume that only bonds issued by a single UK Development Bank can be bought. This is in fact how the Juncker-Draghi Plan works, since the ECB and EIB are both pan-EU institutions backed by the taxpayers of all countries in the union. Why would instructing the Bank of England to buy bonds issued by a UK Development Bank be a problem?

The key difference is that the Juncker-Draghi Plan does not involve "instructing" the ECB to do anything. Indeed no fiscal authority in the EU has the authority to do so. Rather, the fiscal plan is created to take advantage of the ECB's existing monetary policy plans. This is "monetary dominance"; the fiscal authority responds to the monetary authority. In contrast, Richard's plan would force the Bank of England to CHANGE its monetary policy stance in order to do the bidding of the fiscal authority. This is "fiscal dominance", and it would mean the end of operational independence for the Bank of England. It's quite a problem, considering that the UK is a member of the EU, which enshrines the independence of central banks in treaty directives, and the Bank of England is a member of the Eurosystem and therefore (in theory) answerable to no-one.

To be sure, I have pointed out before that "independence" for a central bank is an illusion: central banks are only as independent as politicians allow them to be. So the loss of independence is perhaps not the main problem. Far more important is the fact that if the fiscal authority forces the central bank to purchase bonds without regard to the monetary conditions in the economy, the central bank can no longer control inflation. This was demonstrated by Sargeant & Wallace in one of my favourite academic papers, "Some Unpleasant Monetarist Arithmetic".

Richard recognises the potentially inflationary impact of such a programme:
So long as e-printing money to pay for investment is kept at sensible levels to prevent the risk of serious inflation this process of green quantitative easing could be used to fund the investment Scotland wants and badly needs without risk. 
But this means that responsibility for controlling inflation would no longer rest with the central bank. It would rest with the government agency issuing the bonds. As long as the Bank of England was expected to underwrite infrastructure bond issuance, the path of inflation would be determined not by short-term interest rates but by the nominal value of bonds issued. This would be a fundamental change in the conduct of monetary policy. Indeed it would be hard to see how there could be any such thing as "monetary policy". Monetary conditions would be principally determined by the political imperative to invest in infrastructure.

And this creates a further problem. QE is done on the basis that it can be reversed, so that the effect of the monetary expansion is - at least in theory - temporary. Richard's proposal, however, envisages that purchased bonds would be cancelled. This would therefore be permanent expansion of the money supply. Some consideration needs to be given to how monetary conditions could be tightened if inflation started to rise, since the Bank of England would lack the means or the authority to sell the bonds it had purchased.

Currently, the Bank of England has no plans to do more QE, so it is not possible to construct anything remotely like the Juncker-Draghi plan. Richard thinks the Bank of England is wrong:
And let’s be clear: we printed £375 billion to pay for quantitative easing to bail out the financial markets in the three years from 2009 to 2012 and the outcome is that we now have zero per cent inflation, which is less than anyone thinks economically desirable. So we do in fact need new quantitative easing now to create the new money the economy badly needs to create the moderate inflation that keeps the economy on an even keel.
But sorry, Richard, if this is what you think then you need to influence the views of those on the MPC, not try to end the Bank of England's responsibility for monetary policy.

In short, Richard's proposal is for outright monetisation of sovereign debt. I have argued that monetisation of existing debt when growth is stagnant and both inflation and interest rates are negative and expected to remain so carries little inflationary risk. But the UK is not Japan, and this proposal monetises new debt, not existing debt. I'm worried about the replacement of the nominal interest rate anchor with a "quantity of bonds" constraint that is subject to fiscal dominance, and the lack of any mechanism for tightening monetary conditions if inflation started to rise. Maybe it's just me, but I can't help thinking this would not end well.


  1. Couldn't they control inflation by eg raising taxes?

    1. They could, but how would you determine the general level of taxation required to offset monetisation of ring-fenced sectoral bonds? Also, how would the general public react to taxes being raised not to fund government spending but to control inflation?

    2. Especially if they have been told that the infrastructure programme will cost them nothing because it will all be paid for by the Bank of England.....

    3. But do taxes fund government? Britain has a sovereign currency so the government is the sole issuer of the currency. Government spends money into the economy and then taxes it back to stop inflation. Hasn't that always been the case since the government stopped being a user of gold in 1971?
      The Bank of North Dakota seems to be doing rather well and the Canadians did much better when they issued their own debt free money as far back during the time of Gerry McGeer all the way up to the early 70's.

    4. I'm afraid this is not quite true. You have been reading too much MMT. Technically, yes, government spending precedes taxation. Taxation extinguishes the debt obligations created by government spending, obligations which carry interest rates that can be controlled by the central bank. The nominal anchor is not tax rates, it is interest rates. It could be tax rates, but interest rates are more efficient since they are effective more quickly and are more difficult to avoid.

      There are also good political reasons for not interfering with the nominal anchor. The public is used to interest rates being used to control inflation. It is not used to tax rates being used for this purpose, and it could be hugely unpopular. This too would tend to make tax rates less effective as an inflation control mechanism, especially close to an election.

  2. There’s a fundamental flaw in green QE. This flaw is not central to Frances’s points above, so this comment of mine is a bit off topic, but anyway it’s as follows.

    QE is a form of stimulus, whether it’s conventional QE or QE directed at green investments or infrastructure. And the AMOUNT of stimulus required next year or in three years time is very unpredictable: possibly we’ll need none at all.

    That means that if green and/or infrastructure investment is to a significant extent funded by QE, then spending on that investment may gyrate in a manner that leads to near chaos.

    In general, stimulus spending should be spread as widely as possible, so that when the amount of stimulus changes, the change to SPECIFIC TYPES of spending gyrates as LITTLE as possible.

    And I’m not arguing that we don’t need green and/or more infrastructure investment. I’m saying that that investment should be funded in traditional ways (e.g. a mix of tax and borrowing), with a small increase in that spending (perhaps funded by QE) when stimulus is needed.

    1. This is an important point. Monetary policy is intended to be even across the whole UK, not directed at specific sectors. If monetary policy were deliberately focused on supporting the construction, energy and transport industries - which is what this proposal would support, mainly - then it could also be focused on anything that takes a politician's fancy. Apart from the obvious pork-belly problem, this would cause considerable risk to financial stability as well as price stability, I think.

    2. During the monetary reform debate in the House of Commons Michael Meecher MP stated that only 16% of credit is directed at business lending and credit cards. the rest is speculative finance and housing. Our current monetary system already targets the pork belly in the City of London. So much so where I live in west Wales our per cap GDP is only £15k compared to £71 for central London. At the moment commercial banks target credit lines on what they "fancy". Britain doesn't have a Raiffeisen banking system as in Germany. There doesn't seem to be much price stability in the UK housing market. Remember when the Case/Schiller US house price index was going to be weighted into the LBS stats for inflation? Whatever happened to that proposal?

    3. Frances,

      I agree with your point that stimulus should be "even across the whole UK". Actually therein lies a big weakness in monetary policy, seems to me. That is, interest rate adjustments only influence LOAN BASED economic activity, while not directly influencing activity not dependent on loans. In fact interest rate adjustments don't even influence one type of loan based activity: where a person or firm has borrowed on a long term FIXED rate.

      I conclude that all else equal, fiscal stimulus "spread evenly across the economy" is better than monetary stimulus.

    4. The sheer uneven-ness of Chinese development since around the late '90s is a very large reason why Chinese authorities today have a great deal of trouble getting money to where they see is needed. The big issues is not that the narrow stimulus gyrates, it's what people do to insure against/control that gyration, namely attempt to seize control of the money spout (or obfuscate where money is actually going).

  3. For an accountant, it seems rather surprising to me that Richard Murphy doesn't seem to understand a simple asset/liability provision on a balance sheet.

    His version of "QE" is simply monetarism. Printing money and then spending it. He thinks it is possible to simply remove the liability portion of the balance sheet ("cancelling debt") with no effect on the asset side - magically leaving the UK with an endless supply of cost-free money to spend on infrastructure.

    1. Richard is consolidating the balance sheets of the government and the central bank. When you do this, central bank holdings of government debt "disappear", as he says. The UK's Whole Government Accounts do consolidate the central bank into the government balance sheet. So he has a point. But debt that "disappears" due to accounting consolidation does not disappear in reality. Gilts purchased by the Bank of England as part of a QE programme or in the course of normal monetary operations are not cancelled, they are held. They can be sold or lent to the private sector. it is this that Richard wishes to change.

      Richard wishes the Bank of England to buy back specific bonds not as part of monetary operations (though Richard clearly disagrees with the MPC's current stance) but to support fiscal expansion. These purchases would be irrevocable, since government would cancel these bonds when bought by the Bank. As I pointed out in the post, this destroys the Bank's independent ability to control inflation. It is monetary dominance.

  4. I think there is a big delusion in your assumption (not only in yours): "Central bank controls inflation."
    This is wrong. Central bank controls the interest rate. And with this, central bank can affect the business cycle. Affect, not control. Central bank can raise interest rate and so slow down economic activity. Stimulate is much more complicate.
    Inflation occurs, if we have a wage/price spiral. So it is the wage policy, that controls inflation.
    I think this is the main delusion in modern economic thought.
    Christoph Stein

  5. Taking a side look at this. Would it not be more sensible for the BoE to use the assets it has acquired under QE to capitalise investments banks? I would like to see regional and city investment banks established personally. So why can't the BoE sell back a percentage of its £375bn of assets into the marketplace and use the cash generated to capitalise the new investment banks?

    One of the points about QE was that the BoE could remove liquidity it had created from the marketplace by selling the gilts back. It seems elementally more sensible to use the existing unconventional pot to achieve more productive outcomes.

    1. Bank of England selling its QE gilts wouldn't give it cash to capitalise banks - the money received is destroyed. The Bank can create cash any time it likes, but for the Bank to capitalise banks directly would mean that it would be the effective owner of those banks. Your proposal is thus to replace gilts with 100% equity stakes in new investment banks. As investment banks are always leveraged, and any losses incurred by these banks would be borne by the Bank, this is in effect sinply another form of monetary financing of fiscal deficits.

  6. I wonder if you have any thoughts, given all you say, on why Murphy bothers to add this problematic 'innovative financing mechanism' to his policy.

    After all, as you say, central bank independence is largely an illusion anyway. Don't the DMO and the BoE work together to make sure the rates on (conventional) gilts don't move too far off the policy rate and create an arbitrage opportunity? And if so, why couldn't Murphy's infrastructure investments just be funded through normal Treasury debt mechanisms, with the background knowledge that the BoE is always there to ensure the 'long-term sustainability' of the debt by keeping a handle on the rates?

    1. Essentially, this is my argument. There is more than enough demand from UK residents for safe saving vehicles to support a much higher level of UK government debt without a significant impact on interest rates, and channelling savings into long-term investment in this manner creates little risk of inflation. Therefore I don't see any need for the Bank of England to buy these bonds. All that is needed is for the Bank to stand ready to support the prices if necessary by acting as "dealer of last resort", which it already does for conventional government debt anyway. As I see it, the real problem is the arbitrary political limits being placed on government debt issuance regardless of its purpose and the real rates of return on the associated investments.

    2. Very well explained. Hopefully the point gets through to RM.

  7. Only read this today. Found it very helpful. Thanks a lot.

  8. Frank van Lerven20 August 2015 at 16:04

    A very nice post...Thanks again Frances

  9. "But this means that responsibility for controlling inflation would no longer rest with the central bank."

    This would be a good thing! The central bank can only vary interest rates and interest rates don't, by themslves, determine inflation. It is not reasonable to expect the BoE to be responsible for too much. In addition the BoE is expected to be responsible for more than just inflation which compounds the problem.

    The economy needs to be controlled by a single entity which has control over both monetary and fiscal policy and that entity can only be government.


Post a Comment

Popular posts from this blog

WASPI Campaign's legal action is morally wrong


The foolish Samaritan