Issues with “Issues in the Design of Fiscal Policy Rules”

Bill Mitchell drew my attention to a paper that has purportedly influenced the economic policies of the UK Labour Party. This is Discussion Paper No. 429 from the National Institute of Economic and Social Research – Issues in the Design of Fiscal Policy Rules – written by the then Director of the NIESRJonathan Portes  and the well-known Oxford economist and blogger Simon Wren-Lewis.

Before I examine this paper, please permit me a prelude.

We are all very familiar with the common trope “glass half empty/glass half full”. If you think the glass of water is half empty, you are a pessimist; if you think it half full, you are an optimist. Which one are you?

I have long thought that there are two problems with this “insight”.

The first is that it excludes any alternatives, if you are not one, then you are necessarily the other. When viewed this way, one should look for other alternatives, in this case there is (at least) one. I say ‘at least’, because a single other alternative is all that is required to show this is not a real dichotomy. The realist view is that it is just half a glass of water, with reference to being half full or half empty adding nothing. Now it appears the dichotomy is false, so if you reject one view, it does not automatically follow that you endorse the opposite.

The second issue expands on the realist view in showing is that this is simply the wrong way to look at a glass of water. If I am thirsty the question is: is the half glass of water sufficient to satiate my thirst? Maybe it is, maybe not.  If I am in a country with near deficient water issues, they do not leave the tap running when washing their teeth but, often, use a partly filled glass of water. In that case, a mother might tell off her child for using a glass of water that is half full – now that means it is too much and more empty would be better. Nothing to do with being an optimist or pessimist. In other words, in order to establish the relevance of a half glass of water, context is required.

These two points of a false dichotomy and context are sufficient tools to examine this paper. You will be relieved to know that no great maths skills nor economic training required. Let us proceed.

The key question the paper seeks to address is

there was no equivalent for fiscal policy of the Taylor rule for monetary policy: no simple rule to guide fiscal policymakers…This paper is about the search for such a rule.

The Taylor rule is a simple algorithm to guide Central Banks in setting their interest rate to manage inflation, which is the primary tool for a nation’s Monetary Policy.  In reality Central Banks do not blindly implement this rule, although in the type of economic models that JP and SWL colleagues use, this is often used as a  ‘simplifying assumption’ when modelling Central Bank policy settings.  To the degree that Central Bankers do use the Taylor Rule as guidance that might be a reasonable simplification in modelling an economy, or maybe not.

Ok then, this paper is about fiscal rules. Now there are a few approaches to this, one I am very familiar with is Abba Lerner’s Functional Finance. Now, whilst I had zero expectation that SWL and JP would be advocating anything like this, what I did expect in something called a ‘paper’ on this very topic, was an examination of alternative rules and why they do not apply compared to the rule that the paper’s authors argue for. However there was no even a mention of this (or others), if only to be dismissed by, say, referencing another paper which did the presumed actual refutation. Nothing. In the whole paper. Nada.

Now I have read many papers in many fields in my time, including physics, biology, cognitive science, quantitative finance, philosophy, public health, paranormal studies and alternative medicine amongst others. Without any attempt at some coverage of views on this and arguments in favour of their approach, this paper is already beginning to look like the papers in the certain fields that are full of junk science and pseudo-scientific arguments.

This concern is further confirmed when I read in the second paragraph of the introduction that

basic theory suggests that fiscal policy actions should be very different when monetary policy is constrained in a fundamental way, while the reverse is not in general the case. There are two major examples of where this will be true. The first is when interest rates are at the zero lower bound”[My emphasis].

Basic theory”? “Basic Theory”!!

Now one cannot make all the arguments for a theory to apply to it a particular problem in every paper. That would be an absurd demand. However having studied various macro-economic models such as VARs, ISLM, AD/AS, DSGEs, CGEs and SFCs, to create a hidden or implicit presupposition that their approach is so far from being debatable that it is “basic” is very, very dubious. I knew in advance that  they were never going to argue for it, but to call it “basic” is quite underhand but would, of course, be accepted as standard gospel within the echo chamber of fellow orthodox economists and their acolytes or dupes (if Bill Mitchell’s concern re the Labour Party is correct) in political parties they are addressing. (I have concerns over the Zero Lower Bound argument but will not address that further in this post)

The second is where a country is part of a monetary union or a fixed exchange rate regime.

Here I can agree. The currency regime of a Eurozone economy is quite different to that if the UK’s. However my patience in examining their later policy analysis as a result of their core model will have run out long before I get to that and I will not discuss this any further here.

They complete their introduction, by noting the proposal of an optimum fiscal rule has to, most of the time, be ameliorated by managing “non-benevolent” government’s fiscal policies creating a “deficit-bias”. That is one might have to diverge from the rule to explicitly prevent profligacy and so on. However this is beginning to look like that false dichotomy is discussed in my prelude.

Everything they are doing is to preclude seeing what alternatives there are to their approach. The subtle (but not so subtle later) is that a government is either benevolent or not with respect to deficits and so, therefore “deficit-bias”. Is this the correct thing to be looking at in the first place? Where is the context, productivity (never mentioned), unemployment (mentioned once in passing), inflation – ok that is mentioned 18 times in this paper, you are beginning to see a bias – in the authors.

This post is beginning to turn into a fisk of their paper but I have neither the inclination nor the time to do that for the whole paper, nor  do I think would my readers have the patience! I am going to instead complete my analysis of section 2 on Optimum Fiscal Policy.

Suppose social welfare declines as taxes rise, because taxes are distortionary.

(They discuss a 1% from 20% and 50% base – the 50% base being worse in their view – but that is really irrelevant here). Well I would like to see their definition of social welfare, it is a key concept mentioned 6 times but with no explanation.

I can easily imagine two scenarios where a tax increase can increase social welfare.  The first is when the nation has demand  pull inflation – too much money chasing too few goods – in which case a tax increase reduces aggregate disposable income and reduces demand-driven inflation pressure. The second is with a nation around full employment and there are not sufficient real resources for the government to fulfill its elected social welfare agenda, a tax increase reduces demand – for real resources – and so frees up those resources – people, goods and services – that can be employed by the government to achieve those goals. In both cases, a tax increase serves to increase social welfare.

Now one can posit various objections to my points such as the lead/lag relationships with respect to inflation – but this same objection can also be made with respect to monetary policy with interest change lag effects on inflation. The point I am making is that they are again making assumptions, with many hidden presuppositions,  leaving no room for analysis or debate. A key ever-present, at least in my view, tacit dichotomy implying that if you reject these arguments you must be for non-benevolent government;s and their “deficit-bias”. In fact I am waiting for them to establish that their way of looking at this problem is the correct way to look at it. As I proceed, it appears I am waiting in vain.

They then derive or, rather mostly state but do not argue for, a  straightforward difference equation to minimise as their basic optimum fiscal rule. One that any  A-Level Maths student should be able to comprehend. I understand that many readers may have not studied A-Level Maths but do appreciate that this is not even of university level difficulty. (OK, we might not teach Lagrange multipliers – a technique to help solve minimisation problems – at A-Level but the maths itself, we do)

My unplanned fisking of this paper will be finished when I have completed analysis of this equation.

They assume in the multiplier that there is a government budget constraint not to default on its debt. Well that can come out of a simple balance sheet analysis, as a constraint they make it appear it is of a behavioural consequence but the whole point of their analysis is that it is not.

They examine the interest rate as a cost on the Government, but fail to consider that the interest rate channel is an income source to the private sector, in addition the primary budget balance (which excludes government debt interest payments), both increase private sector sterling reserves – which are not discussed. Of  course, they are fully aware of the difference between the primary budget balance and the fiscal balance (which includes transfer payments – payments not exchanged for goods and services – such as debt interest payments)

They do not explain how the  government can control their deficits, they assume that a benevolent Government can – although they do pay some lip service to automatic stabilizers but these are not in their model! They acknowledge that plans can be disrupted by shocks but that is a woefully deficient approach.

They note that the debt is a stock and the deficit is a flow. However they fail to develop a proper stock-flow view on the relation between Government and Non-Government – the Private and International sectors together.  Their model assumes away by not even considering the level of control that the Government can have on its deficit. There is no  discussion on the effect that the savings desires of the private sector can have on the deficit. No discussion on the Government – Non Government financial flows. No discussion on how trade and capital flows and real terms of trade can affect the Government deficit. No discussion on the goals of government policy that affects the deficit – reducing unemployment, increasing productivity, improving real terms of trade or any such like – only GDP and inflation are discussed – in other words the social welfare considerations that premised the motivation for this paper.

There is scant empirical evidence provided for supporting their view on managing Government Debt or even what it is. Should we just consider outstanding Treasury Issued Gilts or the net outstanding liabilities of the Whole Government – there is zero mention of Japan and its debt/gdp ratios.

However when they end up taking one version of their minimisation process to an, admittedly by them extreme, they say

If 𝛽(1 + 𝑟) > 1, then we get a very different and equally surprising result. Taxes gradually fall over time, until they eventually decline to zero. How can this happen, given that the government has spending to finance? The answer is that debt gradually declines to zero, and then the government starts to build up assets. Eventually it has enough assets that it can finance all its spending from the interest on those assets, and so taxes can be completely eliminated.

So what they are saying is that the private sector becomes in debt to the Government, after all if the Government can perform all its services through assets they own and the revenue they generate, then these assets can only come from the private sector, since, on the pain of double entry accounting, they are also liabilities of the private sector. The government does not tax us, partly since none of us have any ‘money’  issued by the government to pay its tax in! That is the previous now retired government debt (or its net financial liabilities)  were financial assets, issued by the Government to purchase goods and services from the private sector, that had not yet been reclaimed in tax – that is equivalent to all taxes not yet paid – these, on the pain of double entry accounting, were our assets and they have now all been paid. How does the Government build up assets? Somehow we have to pay the government taxes even though there are no longer any reserves from which to pay them and the Government continues (presumably it was already doing this before ‘s debt was retired) to buy our assets – presumably from bank issued credit? But there are no reserves so how do banks settle!  I could go but lets stop! (A proper analysis of is really needed but is outside the scope of my critique, I was only being very simplistic and brief in my comments just above).

Now it is one thing to develop a model and realise it has unusual and surprising consequences. These might very well be worthwhile pursuing and understanding. However in this case, even though it is an extreme outcome, it should tell the modeller there is something wrong with their model.

There may be some interesting things they say in the following sections but I cannot accept their model to use as a basis to consider any of that. Further they immediately provide add a  number  ad hoc assumptions – to the model  (not the policy arguments which follow) which makes this whole enterprise initially based on a clearly stated model, as far as I can see, completely pointless.

So are most everyone is a pessimist about government debt, and everyone should be concerned about a government’s deficit bias. Otherwise you are, or believe or want to be,  beneficiary of a non-benevolent government’s deficit bias.

And it is here I have to stop. There is much more I could say but I am out of time. For further critique of these please see Bill Mitchell’s posts, the one noted at the beginning of this post, one published today and one to come tomorrow.

The whole framing of this discussion is entirely misleading. Where is the context? What is the purpose of Government? What is social welfare? What are the macroeconomic implications, measures and tools? How do the price level (inflation), unemployment, productivity, poverty, terms of trade, taxes and other more specific concerns relate to social welfare. When looked this way the concerns of deficits and debts are secondary, not be ignored but not be the drivers either. Does it matter if a fiscal committee take over driving from politicians, if both are trying to drive by steering with the rear view mirror?!


One thought on “Issues with “Issues in the Design of Fiscal Policy Rules””

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s