Winner of the New Statesman SPERI Prize in Political Economy 2016

Friday 27 April 2018

Macroeconomic Policy Reform the IPPR way

Monetary and fiscal policy makers in the UK seem to think they had a good recession. You can tell that because neither group seem particularly interested in learning any lessons. This is despite the fact that we had the deepest recession since the 1930s, and the slowest recovery for centuries. It is also despite the fact that the level of UK GDP is almost 20% below the level it would be if it had followed pre-recessions trends, and all previous recessions have had the economy catch up with that trend.

You can tell from this paragraph that I do think serious changes are required to how monetary and fiscal policy are done. So does the IPPR, and their detailed analysis and proposals are set out in a new report by Alfie Stirling. The analysis is not too technical, well presented, well researched and I agree with a great deal of what is said. I will look a monetary policy first, and then fiscal policy.

What the Great Recession showed us (although many macroeconomists already knew) is that once nominal interest rates hit their effective lower bound (ELB) [1], monetary policy makers lose their reliable means of combating a recession. The report is dubious about Quantitative Easing (QE) for much the same reason that I have been for anything other than a last resort instrument. In brief, the impact of QE is very uncertain because it is not routinely used, and in addition there may be important non-linearities. It is not a reliable alternative to interest rates.

The report makes much the same point about negative nominal interest rates: partially or perhaps fully removing the lower bound. To quote:
“Like QE, the impacts of negative rates are uncertain and, depending on the behavioural response from banks and savers, could actually reduce spending in the economy, or else increase the number of risky loans (see for example Eggertsson, Juelsrud and Wold 2017).”

I know some macroeconomists will disagree with that assessment, but I think the point is valid.

The report also rejects helicopter money as a solution to the ELB problem. Here I found their discussion less convincing, but they do recognise that a form of helicopter money has already been undertaken by some central banks through creating money to change the relationship between borrowing and lending rates, a point that Eric Lonergan has stressed.

The two reforms to monetary policy that have been suggested and which the report does support are adopting unemployment or nominal GDP as either a second target or as an intermediate target, and raising the inflation target by one or two percent. I have argued strongly for a dual mandate and also for using nominal GDP as an intermediate target, so I have no objections here.

The report recognises, however, that none of their proposed reforms to monetary policy eliminates the ELB problem completely. We have, inevitably, to think about the other reliable and effective instrument that we have to stimulate aggregate demand: fiscal policy. Their proposed fiscal rule is very similar to Labour’s fiscal credibility rule. It includes (a) a ‘knockout’ to switch to fiscal expansion if interest rates reach their ELB, (b) 5 year rolling target for a zero current balance (c) a 5 year rolling target for public investment (d) a similar target for debt to GDP. The last in this list you will not find in Portes and Wren-Lewis, in essence because it involves double counting, and debt targets are less robust to shocks than deficit targets.

If governments followed this fiscal rule, then the ELB would not be the serious problem that it is, because reliable fiscal stimulus would replace reliable monetary stimulus at the ELB. But the IPPR worry that governments might not do what the fiscal rule, and with the knockout what the Bank of England, tells them to do. They are concerned that what they call ‘surplus bias’ might be so strong that the government would not run the deficits that the Bank asks them to run.

To overcome this concern, they suggest an alternative to QE at the ELB: the Bank should create reserves to fund projects that are part of a National Investment Bank (NIB). The NIB would be independent of government in terms of the projects it funded (but not its high level mandate), and it would normally raise funds in the open market. (This makes it different from proposals that the NIB be entirely funded by the Bank: see here.) In an ELB recession, the Bank of England would ask the NIB to fund additional projects, with the Bank providing the finance.

As public investment is particularly effective as a countercyclical tool if undertaken immediately, and as it is usually possible to some degree to bring forward investment projects, this proposal seems a superior alternative to QE, as long as the link between additional purchases of NIB debt and additional investment by the NIB was reasonably clear. The key point here is that although conventional QE might try to stimulate private investment by reducing firm borrowing costs, in a situation where there is chronic lack of demand that can be like trying to push on a string. The same problem should not arise with an NIB. In that sense it just seems like a good idea.

Whether it would be enough alone to circumvent the problem of a rabid surplus bias government during a recession I doubt. The kind of public investment that is easy to ramp up quickly in a recession are things like flood defences or filling holes in the road, rather than the kind of things an NIB would fund. A government suffering strong surplus bias could cut these things quicker than an NIB could fund additional projects. Some form of QE would be more powerful in this respect. The danger in either case is that you just encourage the government to try and get down debt even faster: if QE gives money directly to people, the government just raises VAT.

How seriously should we worry about (design policy for) a government offsetting everything the Bank is able to do to stimulate demand in a recession? The answer may be given by imagining the following scenario. The government operates a fiscal rule that has an explicit ELB knockout. The Bank of England, when rates hit the ELB, requests the government undertake fiscal stimulus. If Cameron/Osborne had been faced with both those things, would they have still cut back public investment? I suspect the answer is no. That of course by implication means that central bankers in Europe played a large part in facilitating (or encouraging) austerity, which in the UK stemmed from a failure to admit the problems of the ELB because of a naive faith in QE.

Which brings us to central bank independence and what I call the conventional assignment (outwith the ELB, monetary policy deals with macroeconomic stabilisation). The IPPR stay with the mainstream macroeconomic consensus in wanting to keep both. People with a more MMT type view, like Richard Murphy, would reverse [2] the conventional assignment, and have fiscal policy doing the macroeconomic stabilisation. I have written a great deal on the distinction and will not repeat that here. However it is worth making one point on independence.

The reasons for making central banks independent are not peculiar to monetary policy. They are that if the complex task of macroeconomic stabilisation is left in the hands of politicians who get secret advice, they can mess things up for political ends. [3] Messing things up can be minor (e.g. delaying necessary measures), structural (e.g. time inconsistency) or explosive (e.g. hyperinflation). Austerity shows that this fear is justified. MMT’s answer to the IPPRs concern about a surplus bias government is that this is just a cost of democracy or the good guys would always be in power, which I suspect many might not find reassuring. Yet that is also why European central bank’s encouragement of austerity was far from helpful to the case for the delegation of macroeconomic stabilisation.

[1] 'Effective' because in practice it is up to the central bank to decide at what point they cannot reduce nominal rates further. 

[2] Not strictly true. In the conventional assignment monetary policy does inflation/aggregate demand and government looks after its debt, while in MMT fiscal does inflation/aggregate demand and government debt looks after itself.

[3] I hope time inconsistency can be subsumed under this broad definition.


  1. Do have sympathy with MMT in that if there is no UBI why should a system be accepting of anyone being unemployed?

  2. As you say this seems a sensible framework in general terms and an improvement on what we have.However a couple of points strike me.

    For a start there is an air of "clockwork" to reaching the ELB and then fiscal policy taking over which doesn't seem realistic. This may not be the trivial point it seems as delays in starting spending undermine the core objective of the policy.

    I think a more serious objection is the five year balance rule which you use in relation to the Cameron/Osborne austerity and I can't understand your view on that. Their objective was to shrink the State and, although the policy was inept and wrong it is a valid political and economic objective.However, it is a policy which implies a long run approach to spending and fiscal policy generally which may not fit easily within a mechanism that is purely counter cyclical with the implied added result of zero addition to public debt (investment is assumed as at least cash neutral).

  3. As a non economist, I would like to know if there is any respectable school of economics which does support the sort of austerity policies followed since 2010 ? Would the economics part of David Cameron's PPE have included any thing justifying pro cyclic economic policies?

  4. The IPPR’s NIB idea is a dog’s dinner. It’s bad enough having two bodies with a say on stimulus (central bank and politicians / treasury), but a THIRD one (NIB) makes matter worse. This is a gross contravention of the Tinbergen principle, namely that for each policy objective, one policy instrument is needed, and one only.

    Moreover, under an IPPR regime, assuming no stimulus is needed at all, then investment spending stops altogether! At the very least, and given the gyrations in the amount of stimulus needed and the erratic behavior of politicians, the gyrations in NIB investment spending would mean poor value for money is obtained from investment spending.

    Politicians have an absolute right to make strictly political decisions, like what % of GDP goes to public spending, but given their erratic behavior (referred to by SW-L and the IPPR) I suggest politicians should have NO SAY on the SIZE of a stimulus package. That should be left to economists, and that’s the arrangement advocated by the submission made to Vickers by Positive Money, the NEF and Richard Werner:

  5. I completely agree with your analysis. The bank is required to invest in 'socially and economically advantageous projects' but at the same time is required to increase credit in the depths of a recession. How can it do this without sustaining losses? Where is the credit demand coming from? How does the bank make sure the stimulus is delivered as soon as it is required (as if the bank could immediately lend hundreds of billion almost over night at the onset of a recession). It seems much quicker and powerful to deliver stimulus through traditional public spending.

    A simple (and explicit) ELB knockout makes much more sense without the added complexity. "Surplus bias" may still prevail but I do not think many governments would want to be seen to resist the advice of the central bank and therefore be responsible for worsening a downturn.

  6. The authors argue that the BoE are conducting "opaque monetary finance" through the TFS scheme. I think they do not understand monetary finance.

    Unless I am mistaken monetary finance is perpetual whereas TFS funds are not. Banks will need to finance any new lending from the market once TFS funding matures.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.