22
Examination of witnesses
Dr Xavier Debrun and Dr Ethan Ilzetzki.
Q30 The Chair: Welcome to our witnesses for our second evidence session—is it our second or our third? It is more than our second, anyway—in our most recent inquiry into the UK’s fiscal framework. We are delighted to have two expert academics on this topic: Dr Xavier Debrun, who is the head of economics and research at the National Bank of Belgium and who is joining us online; and Dr Ethan Ilzetzki, who is an associate professor in economics at the London School of Economics. I thank you both for your time; we are really grateful.
We are broadcasting live on parliamentlive.tv. We will make sure that a full transcript is taken and made available to you so that you can make any corrections to the record after the meeting. I shall start by asking Dr Debrun a question; we will then go over to Ethan Ilzetzki. Can you set out the thinking that led to the creation of a wave of fiscal councils embedded in fiscal frameworks? What was the economic logic of that in the first place?
Dr Xavier Debrun: First, there is a strong relationship between what happened on the monetary policy side and what happened on the fiscal policy side.
You might remember that, back in the late 1970s, there was a whole literature questioning the fact that full discretion was in the hands of policymakers. By “full discretion”, I mean the freedom to do whatever you want whenever you want, regardless of the circumstances or in any circumstances you want—that is, you are completely free to act and to play with the policy instruments in your hands. In the late 1970s, as you might remember, there was a spike in inflation and central banks did not seem to be in control. You had a whole literature developing around the idea of rules instead of discretion. Therefore, there was an impression at the time that monetary policymakers were making big mistakes that led to inflation; and that, therefore, you needed somehow to tie their hands to inflation levels that were seen as reasonable.
In essence, there are two ways to constrain the discretion of policymakers: one way is rules; and the other way is delegation of prerogatives to independent institutions that have a narrowly defined mandate to deliver on some target. On the monetary policy side, we tried rules. At that time, in the 1980s, there was a presumption that you needed to control the money supply. Therefore, central banks were trying to follow rules in terms of the growth of the amount of money that was in circulation in the economy. Those rules completely failed simply because there were many financial innovations in the 1980s, which completely undermined the ability of central banks to control the money supply.
Why do I go to central banks? Because the next step for central banks was, “Okay, we cannot constrain discretion by imposing rules on central banks, so what we are going to do is simply delegate monetary policy to an agency that is independent of the political game. This agency will receive a narrowly defined mandate to target inflation and so on”. That was the birth of independent central banks as we know them.
On the fiscal side, things did not happen like that; they happened with a bit of a delay. In the 1980s, a number of countries—especially in continental Europe; my country is definitely a culprit in that—realised that fiscal policy was not under control. There, the metric was not inflation; it was the fact that there was a steady rise in the ratio between the total public debt and the GDP of the country. If that ratio keeps rising for ever, that is not sustainable. At some point, the budget constraint is going to hit you, and the way in which the budget constraint typically hits you is through a financial crisis; basically, the Government find themselves incapable of accessing financial markets to borrow.
There was a feeling, pretty much like there was with monetary policy, that you somehow needed to constrain the way in which fiscal policy was decided because, apparently—at least in some countries—there was no proper internalisation of the Government’s budget constraint. That is when the idea of fiscal rules started to emerge: “We’re somehow going to impose constraints on deficits and debt”.
The thing that was really a game-changer in the European Union at the time was the fact that there was this grand plan to have a monetary union in Europe. When you have the usual interaction between one central bank and one Government—they can be good or bad; they can lead to stress or whatever—there is only one central bank and one Government. However, in Europe, the Delors committee realised in 1989, “If you have, say, 20 different treasuries interacting with only one central bank, we have a problem”. The problem is that all of these individual treasuries must be well-behaved for monetary policy to continue or to deliver on its mandate.
Of course, at the time, among the group of countries that were about to be admitted in the European monetary union—the euro area—you had diverse relationships, I would say, in terms of different countries and different Governments with the notion of budget constraint. You had countries that were perceived as being very lax and having very high public debt: Belgium, Italy, and so on. You then had other countries such as Germany, which were not only very much focused on the inflation objective but also committed to a significant level of fiscal discipline.
Therefore, those rules were seen as a way to make sure that there was a co-ordination of all the different treasuries around the idea that there were clear limits to deficits and to debts, in order to make sure that the public debt would never become a problem for monetary policy. We can discuss later when public debt can become a problem for monetary policy, but the genesis of the wave of fiscal rules that we saw—certainly in Europe in the 1990s—was definitely this idea that you need to constrain national fiscal policies, to a degree, to standards of fiscal discipline that would be consistent with a single central bank focused exclusively, in the case of the ECB, on delivering inflation of about 2%.
The Chair: That was a very comprehensive answer; thank you. Professor Ilzetzki, do you want to add anything on that?
Dr Ethan Ilzetzki: I fully agree with the broad history that Dr Debrun gave. I would only add that there are other experiences outside the eurozone. There are countries that have had fiscal councils that preceded this. Think about the Congressional Budget Office in the United States and the Parliamentary Budget Office in Canada: these countries do not have formal fiscal rules. There is a separate objective of fiscal councils that relates to fiscal rules but is not necessarily required: simply provide transparency in forecasts. That is desirable, even if you do not have fiscal rules, because the Government have an incentive to provide overly optimistic forecasts. It is good to have an independent body to provide that information.
The Chair: I have a brief follow-up question for both of you; I will start with you, Professor Ilzetzki. Taking on that point, could there be merit in contemplating a fiscal council in the total absence of any fiscal rules? Would that still be worthwhile value-adding in an economy?
Dr Ethan Ilzetzki: I think so. That is not to say that there is no value to fiscal rules, but, if a country did not have fiscal rules, I would still advise it to have an independent body that provides forecasts outside of the political fray. In some countries, that is done by the central bank, so you do not necessarily have to have a new body for that purpose, but having an official or semi-official forecast provided by independent experts is desirable regardless of whether you have a fiscal rule.
The Chair: Briefly, before we move on, Dr Debrun, do you agree with that?
Dr Xavier Debrun: I fully agree. First, it is fine to have independent fiscal councils, but I want to clarify one thing with respect to my earlier statement, which focused only on fiscal rules. On delegation, I spoke about independent central banks and so on, but delegation is obviously impossible in the fiscal realm. This is not something that can happen, simply because fiscal policy is primarily redistributive. As a result, it can only be legitimate if it belongs to elected officials; that is very clear.
You have these independent fiscal councils and independent fiscal institutions, which have only an advisory role. They can be desirable, even without fiscal rules, because transparency is good; indeed, this is what facilitates the task of Parliaments in holding the Executive branch accountable.
In terms of what they should do, the example of the Congressional Budget Office in the US is a good one because it focuses on this transparency job—that is, providing the forecast and telling Governments, “Look, if you don’t do anything over the next 10 years, this is what’s going to happen. This is where public debt will be. This is where the deficit will be”. The CBO is well known for providing these really large numbers calculated over 10 years—it is always tens of trillions of dollars—but this has a value.
Another important role that they play is in the scoring of policy measures. The CBO is doing that. It is obviously valuable information for Parliaments before they can make fully informed decisions. That being said, there are strong complementarities between fiscal rules and fiscal councils, but we can discuss those later.
The Chair: Thank you, both; that is very interesting.
Q31 Lord Petitgas: We are all clear, I think, that we need to enforce fiscal discipline because the temptation of politicians is always to spend more and to try not to tax people—at least, that is the general view. The question now is: who enforces the fiscal discipline and how is it done? The question is about—I think that you, Professor Ilzetzki, wrote about this—whether the fiscal discipline is enforced by the markets. Is it enforced by a council—or an equivalent of the American office—or the OBR, or both? The complication now is the interaction of markets versus the OBR in the UK, as well as the expectation of what it will say and the Government’s reaction to the scoring. Everybody is trying to second-guess. Are the markets looking at the watchdog or is the watchdog also looking at the markets?
Dr Ethan Ilzetzki: We do not want to get to a point where markets are enforcing fiscal discipline because, as Dr Debrun pointed out, that can have a large cost to the economy. The rationale behind fiscal rules is to avoid arriving at a point where markets are providing that discipline. I expect that there might be some differences between myself and Dr Debrun, but I think that the empirical evidence on whether markets truly are using the fiscal rule in any way as a guide-post is extremely mixed and is driven primarily by middle-income, not high-income, countries.
Pointing to suggestive anecdotes is quite useful here. If we look at, say, the case of Greece, Greece had been violating its fiscal rules for a good decade before its crisis. The markets shrugged that off entirely until they did not, but it was not because that was a point at which the rules were violated. The rules were never binding. The markets did not care, then they did care. We look at France today. I believe that France has been violating its fiscal rules since at least 2018; it is in a semi-constitutional crisis over this matter. For whatever reason, the markets are just yawning this off as insignificant, and France can still borrow at among the lowest real rates that any country has ever borrowed in human history.
Markets are rather sophisticated: they can see through manipulations of fiscal rules and the creative accounting that is used to meet the fiscal rules. Also, when fiscal rules are violated, they are wise enough to read the overall economic circumstances and not view that as some necessary signal that the economy is in a crisis. In other words, it is neither a necessary nor a sufficient condition; a fiscal rule was neither a necessary nor a sufficient condition to avoid a financial crisis.
Dr Xavier Debrun: I generally agree. If we talk about the evidence on fiscal rules, it is true that it is quite mixed. The correlations are strong, but it is true that establishing the causality—that is, is it really the presence of a rule that induces a different fiscal behaviour?—has not been settled yet. There is still a debate.
There is one thing I would like to emphasise, which is the difference between enforcement and, ultimately, compliance with fiscal rules and the effectiveness of a fiscal rule. Of course, the legal meaning of compliance is whether you are above 3% or below 3%; if you are above, you do not comply. If you use that metric, and look again at continental Europe and the euro area, you will find that compliance rates are about 50%. They pretty low. This is a common feature of a lot of countries with fiscal rules; formal compliance rates are very low.
At the same time, we have evidence—that evidence is not causal; it is still a question mark—that there is a different behaviour for countries that tend to stick to fiscal rules. It is a bit like when you are driving on a road and there is a speed limit of no more than 50 miles an hour. If you see the sign, you will say, “Okay, I may be driving at 55, but I will not go to 65 because I would feel bad driving at that speed”. If there was no sign, you might as well say, “I am alone on this road and it is straight, so I might as well drive at 70”. Who prevents me doing that?
One of the last papers to which I contributed when I was at the IMF—back in 2018, I think—identified the magnet effect of fiscal rules. If you have a fixed number of 3%, somehow, countries that are subject to this 3% will feel uncomfortable deviating too much. In looking at not just the point estimate of the effect of fiscal rules but the entire distribution of outcomes, we identified that countries that were subject to the 3% fiscal rules had a significantly different distribution of fiscal outcomes than countries that were not subject to that 3% limit. This is what we call the magnet effect. Fiscal rules make it uncomfortable to cheat openly because the public understand what 3% means and that 3.5% does not look right. You would not have that result with a fiscal rule.
I totally agree that you do not want the markets to be the enforcer because markets are fickle; fiscal discipline through markets is extremely brutal, violent and, in the end, costly. What you have in continental Europe is enforcement via a supranational institution, obviously, but there an issue in that you have a big difference between large and small countries. Small countries such as mine tend to listen to what the Commission says but, if you are France, you do not necessarily listen.
I do not think that enforcement through independent fiscal institutions is feasible. I have only one example in mind: Hungary in 2008. It was before the current Prime Minister. It tried to have a tough fiscal framework with a strong fiscal council that had, in effect, through the constitution, a veto on the budget; if the fiscal council considered that the budget was inconsistent with the constitutional principle of fiscal responsibility, the budget could not be tabled to Parliament and the Government had to prepare another budget. Of course, as you can imagine, that constitutional amendment failed because nobody wanted to give such a big power to a body made up of unelected officials.
The Chair: Lord Davies, did you have a brief follow-up question before we move on?
Lord Davies of Brixton: I will come back to it later.
The Chair: Okay. Lord Lamont, you have a follow-up question and are also asking the next question, so you can roll them into one seamless whole if you would like.
Q32 Lord Lamont of Lerwick: I will start with Dr Debrun. What is your assessment of the fiscal rules that we use in the UK? Do they, as some people allege, inhibit countercyclical responses in a downturn—that is, the use of automatic stabilisers and so on? We have also been told by some people that the rules incentivise Governments to operate without a sufficient fiscal buffer—that is, people go right up to the edge. Do you agree with those two points?
Dr Xavier Debrun: To be completely honest with you, I do not feel sufficiently familiar with the UK’s fiscal rules, as they are right now, to make an assessment. However, to follow up a bit on some of your points, on the idea that fiscal rules would somehow prevent Governments or discourage the accumulation of buffers, if that were the case, it would be a poorly designed fiscal rule. The idea is precisely to have fiscal rules to encourage the accumulation of buffers.
What we realise in the empirical literature about fiscal policy is that those countries that have fiscal rules tend to comply with fiscal rules when they are binding, which is in bad times, but, in good times, when the rule is not necessarily binding—especially if you have a deficit rule: in good times, you have a lot of revenues and can therefore comply with your fiscal rule much more easily—since the rule is not binding, Governments do not tend to accumulate sufficient buffers. That is the idea; the rule is not sufficiently binding. However, if you had no rule, Governments may not accumulate those buffers anyway. It is an issue with the fiscal rule’s design.
What they have tried to do in continental Europe is, again, to modify the stability and growth pact and say, “It’s not so much about the 3%, but we need to have a medium-term view and everything needs to be focused on expenditure. Countries need to commit to a medium-term expenditure path”. The idea there is precisely that, if you commit to an expenditure path, when times are good and you get a lot of revenue windfalls, the rule, because it is constraining expenditure, forces you to accumulate more buffers. In principle, that is a well-designed rule; on whether it is going to be implemented, I have my doubts, but I will leave it there.
Dr Ethan Ilzetzki: In the case of the UK, there is nothing inherently wrong with the fiscal rules themselves, but the way they are applied in practice does lead to a slight procyclical bias. Let me explain.
In extreme events—things such as Covid, say—the fiscal rules have escape clauses here; indeed, we saw in Covid that the fiscal rule did not bind. In severe crises, there is still an ability to conduct countercyclical fiscal policy. The problem is more in times such as those we are currently experiencing, where things are not fantastic. We are not officially in a deep recession, but it is at times like this that the fiscal rule becomes heavily binding on policy and somewhat constraining.
To make matters worse, the UK is a prime example of what Dr Debrun referred to—I very much appreciate his research on this matter—in that it is a case of a fiscal magnet. Roughly, 3% was the peak deficit that was occurring before the 2008 crisis; now, it is almost the minimum deficit that Governments are willing to run. The idea is that, when the OBR gives a Government an assessment of fiscal space, no Government in their right mind would leave that fiscal space on the table. There is little incentive for Governments to fix the roof when the sun is out.
This leads to conditions where we find ourselves in a situation, as we are at the moment somewhat, in which some moderately countercyclical policy may be useful but the fiscal rule—I am not talking about some massive expansion. Let me put it differently: we would not want to have erratic increases in taxes every Budget because of small changes in forecasts. This is the current application. Again, nothing in the way the fiscal rules are designed requires this, but, in the way they are applied, because these are medium-term targets, one could imagine a Government choosing to meet that target over time. However, the way the rules are applied in practice are such that any change in a forecast leads to an erratic change in policy.
One more thing is that a negative side effect of the fiscal rules is that they disincentivise deeper structural reform, because they give a fig leaf of fiscal responsibility: “Yes, we are meeting our fiscal target, so everything is okay”, but we are trying to solve what is in essence a structural problem with patchwork in every Budget. The problem is that, structurally, expenditure is higher than taxation in the United Kingdom. What is required is deeper tax reform, deeper expenditure reform or a combination of the two. However, no Government really feel the need to do so because we have the fiscal rules so we appear to be behaving in a fiscally responsible way.
I have one addendum to that. Let us look at a couple of the more successful cases of building fiscal buffers. These are not cases where the fiscal rules are binding or enforcing anything. Greece is currently doing a fantastic job—we will see how long that lasts—in running primary surpluses of 3%. So, the Greek economy is doing well, but, in the past, the Greek economy has done well but the government spent all of the proceeds of that success. Greece has done tax reform and labour market reform, which we hope might bring longer-term fiscal sustainability to Greece. These things are much driven by their experience of being burnt during the crisis.
I do not think we should wish upon ourselves a crisis to bring those reforms, but I feel that there is nothing inherent in the fiscal rules that prevents a Government engaging in deeper reform. The political culture is such that it provides an impression of fiscal responsibility without having to make the more difficult choices in which Governments need to engage.
Lord Lamont of Lerwick: There has been a lot of criticism—or speculation about criticism—in the UK recently that small changes in forecasting in-year can force Governments into abrupt changes of policy. Of course, you can say, “That’s because they’ve used up so much of the headroom”, but if the headroom itself becomes the target—or is, in the minds of Governments, the target—surely you face the same problem again, in that a slight change in forecast can bring about an abrupt change of policy.
Dr Ethan Ilzetzki: Let me say something that is somewhat controversial, if I have not done so already: I would want to dispose of the concept of fiscal headroom entirely.
Ultimately, what does fiscal headroom mean? It means that the Office for Budget Responsibility forecasts. If we say that there is £10 billion of headroom now, it means that, if the Government borrowed £10 billion less, there would be a roughly 50% chance of the Government meeting their medium-term fiscal target. The £10 billion reduces that probability to 48% or 46%. There is no line in the sand that this fiscal space indicates. Although we would want to keep buffers, of course—I would want to see an even greater buffer than £10 billion; I am on the same page on that—an attempt just to scrape a couple of additional billion, to increase that fiscal headroom on the margin, may be as politically expedient so that the Government can say they are meeting their medium-term fiscal commitments. However, it is not enough.
Also, in my view, the erratic policy to increase the probability of meeting a target with enormous margins of error with slightly higher probability does not seem worth it. To put it differently—I will be very concrete here—if the question of whether we should raise national insurance contributions in order to increase the probability of the Government meeting their target by two percentage points were put to the public, I do not think that the public would want to do that.
There is something about how the Office for Budget Responsibility communicates. This is not a criticism of how it operates—there are very wise people in that institution, and they are doing the best they can—but something has gone a bit awry in how its forecasts are applied in practice to the fiscal rules, requiring this knee-jerk response with every budget.
Q33 Lord Lamont of Lerwick: I want to ask Mr Debrun a question relating to his answer to the question prior to mine. I was very interested in what you said about the origin of fiscal rules within the currency union in Europe, but is not the significance of fiscal rules in a currency union completely different from fiscal rules outside a currency union?
For example, some countries in the euro area have worse fiscal positions than Britain’s but none the less enjoy lower 10-year bond rates because are operating within the currency union and they have the resources of both Germany and the ECB. Surely the whole argument about fiscal rules and what is happening in the EU is completely different and not relevant to the debate about individual sovereign countries.
Dr Xavier Debrun: Yes and no. You are completely right that it is different, but it is different for reasons that have to do with the fact that, if you are in a currency union, you have a form of incomplete federalism at play. You are right that, despite quite different fiscal behaviours, the risk premia in the euro area are still extremely compressed; I completely agree with you on that. The question of whether you want to have a fiscal rule is simply about whether you realise at some point that you have a fiscal problem. If you realise that, systematically, your deficit is too high, regardless of whether you are in a currency union, you should probably go for a fiscal rule.
It could also be about strong societal preferences for fiscal discipline. Germany has had a fiscal rule since the very beginning. It did not wait for the Maastricht treaty to have a fiscal rule. Many countries in Latin America, which is not in a currency union, realised at some point in the 1990s that they had a serious fiscal issue, which was leading to difficult relationships with their central banks and, therefore, excessive inflation. They decided to adopt fiscal rules on their own. Switzerland is not in a currency union but it has decided to adopt a fiscal rule—the Swiss debt brake—which is extremely effective. It had a trickle-down effect on the cantons—the federal states in Switzerland—which all adopted fiscal rules that were very similar to those of the federal state. You could see the impact on the interest rates spread. You are right that there are aspects that are totally peculiar to participation in a currency union, but the case for fiscal rules and fiscal councils is a pretty general one.
If you allow me, I will say a couple of things on what was said earlier about small changes in the forecast leading to policy changes. I emphasise that one of the benefits of fiscal policy rules—at least, one of the expected benefits, or one benefit that we would like to see—is increasing the predictability of fiscal policy. The benefits of predictable fiscal policies are large because you help markets form their expectations and you basically lower risk premia; even if it is by only a few basis points, you should have an impact. It is regrettable that the application of a fiscal rule would lead to a form of policy fine tuning, which a fiscal rule should discourage. There, I completely endorse the point made by Dr Ilzetzki.
Q34 Lord Burns: I interpret from you that, in a sense, there are two meanings of headroom. One is how much scope there is to make use of that in the immediate period ahead. The second is how much space you need, in terms of the rule, to allow for the inevitable variations or errors in forecasts. As one goes out in years two, three and four, those errors become too big. One of my worries about what has happened in the UK is that these two notions have become hopelessly confused and led to the view that, if there is a bit of space, you should go out and spend it immediately.
Dr Ethan Ilzetzki: I absolutely agree with your assessment. When the fiscal headspace is communicated to the public, what is in people’s minds is that they are imagining that, if £1 more were borrowed, there would be a financial crisis; obviously, that is not what it is supposed to be communicated. This pressures Governments to respond erratically and ad hoc to the loss of fiscal headspace because of changes in forecasts.
Conversely, as you rightly point out, it disincentivises building fiscal headspace, or additional fiscal headspace, because the political pressures in Parliament are such that, if you get a good forecast and have more fiscal headspace, the impression is, “We have all this money to spend. Why should we not spend it or cut taxes?” Again, something has gone a little awry in how we interpret the notion of fiscal headspace.
Q35 Lord Blackwell: I want to ask about the impact of the OBR in the UK more specifically; I will start with you, Professor Ilzetzki. We can all see that the combination of the fiscal rules and the OBR is intended to create a discipline that avoids excessive debt build-up, the consequent rise in the risk premium on interest rates and so on.
The question is: has the OBR been successful in helping constrain fiscal discipline in the UK? If the answer is yes, the criticism made is that financial forecasts are not completely objective. All economists can agree that, in the long run, the only source of economic growth is productivity, but people will argue that the best source of productivity is low taxes to encourage enterprise and incentives. Other people will say that the right way to encourage productivity is spending on education, training, welfare and government investment in infrastructure. There is great room for debate about what policies will achieve the growth, which ought to be something on which a democratic Government can have a view and exercise their policies. If this is outsourced to the OBR, the criticism is that a group of economists sitting in their office—it is not quite an ivory tower, but an independent office—are constraining the action of the Government to follow their chosen policies. How would you react to all of that?
Dr Ethan Ilzetzki: I have specific criticisms—they are far too technical to be worth bringing up here—of some of the ways in which the OBR has conducted its forecasts, but I will not get into those.
Lord Blackwell: Why not?
Dr Ethan Ilzetzki: Let me begin by saying that the OBR is doing the best it can. I do not think that there is any malice there. It is truly attempting to give the best forecast, and the methodology it uses is not miles away from what the Bank of England and the IMF do. The tools it uses are pretty conventional. However, even those best tools have enormous uncertainty around them.
The truth of the matter is that, five years out, we are going to have three or four shocks—positive or negative—and all of these forecasts are going to be yesterday’s news. There is an enormous amount of uncertainty. The only thing we can say with certainty is that the forecast will be wrong. In which way will it be wrong? I would say that it is about 50:50; that is a good forecast. However, we must treat it as such. We must understand that the margins of error here are really large. It is not so much that we do not want an independent assessment or evaluation; as you point out, assumptions go into that assessment, and not all economists would agree with the assumptions that enter those assessments.
Secondly, it is about the communication of the enormous uncertainty. Let me give an example from the fiscal council in Ireland; this is very much in line with what you were suggesting. The fiscal council there provides its best forecast, but that forecast is not binding on the Government. The Government may choose to declare themselves in compliance with the fiscal rule on the basis of their own forecast.
The cost to the Government of being overly optimistic here—the fiscal council still plays an important role because the risk for the Government—is that, if they are viewing the world through rosy glasses too much, they will have to come to the electorate and explain why they made these entirely unrealistic forecasts, or come to the next budget and explain why they made these unrealistic forecasts and why the fiscal council was correct and they were wrong.
I am not necessarily endorsing that model, but I am giving it as an example of the different ways in which one could use the forecast of the fiscal council. Using this as a single point estimate that is de facto legally binding on the Government is a little problematic. I am happy to get into details if there are follow-up questions.
Lord Blackwell: Are you pointing to the difference between a Government who have outsourced forecasting to the OBR, in effect, and other systems where the Government do their own forecasts and the OBR equivalent is commenting on them or providing an alternative?
Dr Ethan Ilzetzki: Exactly. Again, by no means would I dispose of the OBR. It is important to have these independent forecasts and not to allow politicians to make whatever forecast they wish to make. Of course, that provides some anchor of realism to any forecast. If the Government all of a sudden say, “We think growth is going to be 10% a year”, and the OBR says that it is going to be 0.1% a year, there is some explaining to do in terms of how they came to that conclusion. It provides some anchor of realism to any government forecast but the Government may say, “We believe that our tax cuts are going to be stimulative and, therefore, we think that we are going to be within our fiscal envelope”. They will have to answer to the public if, a couple of years down the line, their forecasts turn out to be wrong.
Lord Blackwell: I will come to Dr Debrun in a minute. If you look at the international comparisons of different fiscal watchdogs, ranging from those where the Government are clearly in control and the watchdog is just there to comment versus what we have in the UK—in essence, the OBR is seen as the arbiter of whether the Government are complying—is the UK at one extreme of that spectrum?
Dr Ethan Ilzetzki: It is at one extreme, but it is not an outlier. There are quite a few countries in which that is the case; it is not unusual to have that.
Let me add to that other models. In Canada, which does not have a fiscal rule, the Parliamentary Budget Office scores the party manifestos. Another challenge that we face in the UK—I should not say “triple lock” because that means something extremely specific—is that I feel like Governments are bound. They have their party manifestos, which are viewed as a semi-contractual agreement with the electorate. They have the fiscal rule, which is viewed as a semi-legal contract with the electorate. They also have the internal political pressures of their parliamentary majority.
In essence, in the current circumstances, it means that you cannot raise taxes because you have promised not to raise taxes; you cannot have a deficit because you have promised not to have a deficit; and, with the risk of getting slightly into politics, you have a parliamentary majority that would refuse to cut public spending. The numbers simply do not add up. We need some leeway here on one of those margins. I would rather it not be on the political side—not talking specifically about this Government—but something has to give.
In Canada, what they have done is score the party manifesto. The electorate can then go and decide how much they want to allow the Government to borrow. There is evidence that Governments are penalised electorally for proposing excessive deficits; the public are not thrilled to see Governments overspend or under-tax.
Lord Blackwell: Dr Debrun, do you want to add anything on the extent to which these fiscal watchdogs in other countries are subject to the same criticism of being overreaching—or, potentially, political, in their view—or is the UK different from other countries?
Dr Xavier Debrun: If you look at other countries, it is correct that, since the forecast is compulsory, this is one extreme of the spectrum. There is no doubt about that. The UK is not alone in that club: Austria, Belgium, and the Netherlands have the same model. Is it subject to criticism? Not in those countries; I have not heard any. Especially in Belgium, nobody has questions about the fact that a bunch of experts sitting in what we still call the Federal Planning Bureau are providing the forecast. This is a bit of a relief for the Ministry of Finance, which is not necessarily staffed with all the experts required to provide a credible forecast.
There are other models. Look at Spain and Italy, for instance. They are intermediate. Spain’s and Italy’s fiscal councils, or the equivalents of the OBR, must validate the Government’s forecast. The Government provide the forecast and ask, “Is it right or wrong?” Then the fiscal council says yes or no. If it says no, something will happen: the forecast will be revised until it is deemed acceptable. There is formal validation in Spain and Italy.
I remember my time at the IMF and talking to Italian and Spanish officials. Of course, this is purely anecdotal evidence—it is not statistical in any way, shape or form—but they would tell you, “Yes, it made a big difference to know that we had the Independent Fiscal Authority in Spain or the Parliamentary Budget Office in Italy watching what government officials were doing with the forecast”.
One clarification I would like to bring in here is that—I heard about forecasts, growth and so on—we need to distinguish between the macroeconomic forecast and the actual fiscal forecast, which is the forecast for revenues and some expenditures that are made. You have a lot more fiscal expertise in the latter, the fiscal forecast, than in the macro forecast. You have a lot of macro forecasts to look at.
Look at any investment bank: they have their micro forecast. Look at rating agencies: they have their micro forecast. Look at the IMF, the European Commission, and so on: they have their own forecasts. It is harder for Governments to cheat or fudge a micro forecast than it is for them to fudge a fiscal forecast. This is where the expertise of independent fiscal institutions is particularly useful. It also holds the Executive branch’s official forecasts a little more accountable if they try to use their margin of expertise on fiscal issues to come up with more rosy revenue forecast than is warranted by the growth assumptions and the growth forecast.
More fundamentally, I was invited to supervise a study for the French Supreme Court of Audit—the Cour des Comptes—which wants to beef up the role of its own OBR in the fiscal forecast. This was precisely the argument. It asked, “Are the forecasts of independent fiscal institutions better?” The answer was, “Not that much”. It is not about the quality of the forecast because, as Dr Ilzetzki reminded us, you are sure to be wrong when you try to do a forecast. The problem is not so much the precision; it is in making sure that the distribution of errors is 50:50. The way we say it in statistics is that we do not want to have any bias; we do not want to have any systematic error in one direction versus another.
In France, all of this work was done by the Cour des Comptes precisely because there were huge forecast errors; they continue to be an issue. These errors led to a perception that there was manipulation by the French Treasury. From talking to French Treasury officials, it was difficult to get to that conclusion; frankly, it was difficult to conclude that there was any manipulation. There were just honest mistakes and errors. However, there is still a broad public perception that the Government are trying to fudge the numbers, which is undermining the quality of the public debate.
If increasing the role of an independent fiscal institution in the forecasting process—it depends on the constitution of each country, the political traditions of each country and so on—in a way that ensures that we find a way for the independent fiscal institution to reduce this perception in the general public, that is really a win. It is a contribution to fiscal policy credibility.
Q36 Lord Turnbull: The problem that is troubling me is referring to the OBR as an independent forecaster. It is a real muddle. There are certain things where the Government tell it what to put in this forecast: “Do not put in anything for the increase in road fuel duty, and take the assumption that total public expenditure will be contained to 1% in real terms”. There are other things where they say they are going to restrict the number of immigrants to this or reduce the number of benefit claimants to that from this measure. Who is deciding that?
What do we have? It is a hybrid. It is not really an independent forecast. Those people would say, “If we were a bit more like the national institute, and if we were actually making a forecast, this is not the forecast we would make”. The public does not really know what to believe. Is this part of government, a compromise between the two of them, or is it really an independent forecast? I suspect it is not.
Dr Ethan Ilzetzki: I do not think, again, the UK would be an outlier in that respect. The way I think the OBR would put it is that it needs to receive the inputs on fiscal policy from the Government and then it is their role to score those policies. I have some understanding of the back and forth that goes on between Treasury and the OBR, but I do not want to dwell on that too much. I do not think it is unusual for there to be an exchange of views between the Treasury and the fiscal bodies.
Lord Turnbull: What does it do if it thinks what it is being asked to cost is implausible?
Dr Ethan Ilzetzki: Do you mean that it is an implausible task to be able to cost it? Is that your question?
Lord Turnbull: This expenditure target—the chance of hitting that is nearly zero if you ask any of them, “Do you think that is what is going to happen or anywhere near what is going to happen?”
Dr Ethan Ilzetzki: Yes. There are certain conventions of how to score policy and those conventions have some very unrealistic assumptions in them, which I guess adds to the question of how much we should think of the point estimate of those forecasts as being—it does de facto mean that the OBR is forced to pretend to be surprised over and over again about policies that—
Lord Turnbull: The combination of these two is the improvement in the fiscal forecasting deficit by 2029-30. Things do not get any worse in that period. Yet one does not have a great deal of confidence: if you were free to make a forecast, that is not the outcome you would come up with.
Dr Ethan Ilzetzki: The margins of error are anyway so large that they swamp the bias that is introduced to the forecast because of the concerns you are raising, which are true concerns. I also worry that we are not doing enough to reduce public debt.
The Chair: Baroness Wolf, did you have a follow-up?
Q37 Baroness Wolf of Dulwich: Yes. I was very interested in what you had to say, Dr Debrun, about current mistrust in France and what is going on, because it leads into my question and follow-up.
Earlier, Professor Ilzetzki referred to the fact that—in common with a number of other European countries—we have a situation where we have a structural deficit and low growth. In that situation, a lot of headroom is something you dream about.
None the less, I think it is true that in this country we and the public have become extremely aware of the Treasury OBR dynamic, which is going to be there with any of these systems. You score it, and then the ministry of finance has to look and move. That is going to be true under the surface anyway. That is up to the point of the whole exercise. However, it has become very visible in this country, and I think that people feel it has become a distraction from a more sensible discussion of what is going on.
My question is: is that related to the structure of our system and the visibility, the forecast, and the fact that they are statutory? Or is that just a function of the fact that we are in a very tight fiscal situation? When you look at other countries, does anybody seem to be less prone to ending up in a situation where the details of the argument swamp the underlying structural realities?
Dr Ethan Ilzetzki: I can compare it to the United States, without a fiscal rule and with a fiscal council that has no binding authority over the budget. It does make news when the Government come and say, “Somehow we are going to cut taxes and that is going to raise revenues” as some politicians occasionally say, and the Congressional Budget Office says, “No. We think that is going to lose revenue”; indeed that does make headlines.
I do think that the obsession in this country with the fiscal space is at an extreme from what I have seen anywhere. I wrote down for myself the headline from BBC Radio 4 on Thursday: “Economic growth has slowed to 0.1%, in a setback to the Chancellor in advance of her Budget”. The setback is to all of us. The setback to the Chancellor of growth being slow is not only because of the budgetary implications. This is bad news for us. Everything is read through this lens of whether this is going to increase fiscal space by £1 billion or reduce it by £1 billion, again, when that number is a meaningless line in the sand. That leads to a difficult policy dynamic.
Allow me to propose an alternative—and this may be going too far in a different direction—if the OBR instead had communicated something more probabilistic. “We think that the Government will meet their fiscal target with a probability of 46%”; that does not sound great either, but it might incentivise greater fiscal responsibility because you would want to push that number up to 75%. At the moment, the Government are making very erratic policy to change the probability of meeting a fiscal target with small percentage points of probability. If that probabilistic communication became more front and centre—if you read the OBR reports, it is hidden somewhere in there—it would help a lot to change how this is communicated publicly.
The Chair: Dr Debrun, do you want to respond to any of this?
Dr Xavier Debrun: Yes. About the mistrust in France, it is not only in France that people have questions about what Governments tell them about forecast or headroom or whatnot.
In general, one of the greatest contributions that independent fiscal institutions can make is to avoid the politicisation of exercises and actions and duties done by administrations that are essentially technical in nature. I remember a debate in Belgium on what elasticity was used between the gross domestic product and the revenues to lead to a certain forecast. I do not think we should have a political conversation around elasticities.
The same happened when we used so-called structural balance rules in the EU, and at one point it was also the case that the UK used the structural balance rules. The only way you can understand that kind of rule or that kind of indicator is by understanding what the output gap is: the difference between the actual output and what you think would be the maximum output that the economy could produce. Nobody has ever seen an output gap. This is a very technical exercise to put together and to put out there. But it is an essential component of your fiscal indicator, and it became a topic of conversation in Parliament.
I do not think that is where the debate should be. It is not about the output gap. It is about the policies that we put in place. I completely agree with Dr Ilzetzki about the fact that the OBR should communicate a probability of reaching a target, which would then provide an incentive to take measures to increase that probability. I will leave it there for now.
The Chair: Baroness Liddell has a quick follow-up question.
Q38 Baroness Liddell of Coatdyke: Yes, a very quick follow-up. It is quite interesting listening to the interchange here because there is an interesting quotation from Andy Haldane, the former chief economist at the Bank of England: ”The creation of the OBR was lauded except in one respect: unlike the practice in other countries it was made responsible for macroeconomic forecasts and assessing the impact of fiscal measures”—and then comes the really interesting bit—“the Treasury’s role and expertise on macro matters was outsourced”. Bearing in mind we have two former Permanent Secretaries and a former Chancellor of the Exchequer here, do you agree with that?
Lord Razzall: Do not take into account the rider. You can say anything you like in front of them.
Dr Ethan Ilzetzki: It is the case—and the UK is no exception in this regard—that Treasuries tend to have less of a technical expertise than central banks do. It is not uncommon for Treasuries to require outsourcing of their forecasts to other bodies. It is okay to get a second opinion. The issue is if that is the only opinion and it is a de facto legally binding opinion. That is where it becomes more challenging.
Q39 Lord Razzall: Both of you have made quite a number of comments about other countries and what happens there by comparison with us. You are both even more familiar than any of us with the structure of the OBR here and our fiscal framework. I would like to try to pin you down on what recommendations you could make for us looking at international frameworks. Looking at the OBR first—you understand what the existing mandate is—do you think we ought to be changing or recommending a change to the OBR mandate? Secondly, you have mentioned a number of other countries. What other countries do you think have a fiscal framework that we should learn from in our recommendations in our report?
Dr Ethan Ilzetzki: I would not take the fiscal framework of any other country off the shelf. Each country has created a fiscal framework to solve its own specific political economy challenge of deficits. So, Chile may have something related to copper prices, which clearly would not be the right idea for the UK. The Netherlands—which I often hear as a positive example—does have a good fiscal framework, but it is intended to solve the specific problem there of coalitions of many small parties, which again is not the problem in the UK. In the UK, the main problem is that, de facto, the majority party is a dictator for five years. The challenge here is to avoid Governments passing on excessive debt to their successors and exploiting their temporary role to cause permanent damage to the economy. With that in mind, to be more concrete on recommendations, there are a few things that I would adopt from other countries, or at least I would investigate adopting. First, as I mentioned before, the office of the Parliamentary Budget Officer in Canada. It scores not only existing policies but also party platforms, manifestos, given how important the manifesto, as a quasi-constitutional—
Lord Razzall: They do that before the election?
Dr Ethan Ilzetzki: They do that before the election, exactly, so the public get to see what the parties have proposed. Unlike Canada, I would not have that in lieu of a fiscal rule, but I think that, in addition to a fiscal rule that would be a useful thing to have.
I suggested earlier, with respect to the Irish Fiscal Advisory Council, the notion that the Government should be allowed to second-guess the OBR. The Government, as was suggested by one of you—
Lord Razzall: They are allowed to here, but they just do not do so.
Dr Ethan Ilzetzki: Yes, exactly. I think to maybe enshrine that a little more formally that there is that ability for the Government. Like Dr Debrun was saying earlier, I do not think it is about different growth forecasts, but if a Government believe that their policy will stimulate growth, they should be allowed to run on that in the next election and pay the price if they did not stimulate growth with that policy.
The third point that I have made, that I would repeat, is regarding probabilistic forecasts.
Lord Razzall: Yes, you said that. That is quite important.
Dr Ethan Ilzetzki: I believe so too.
Dr Xavier Debrun: Empirically, if you look at the different features of all the independent fiscal institutions, like the OBRs, there is a huge amount of heterogeneity among them. The mandates are different: some are nested in Parliament; others are stand-alone institutions; others are nested in the Treasury or in the central bank, so the models are very different. Indeed, they do correspond, as was said by Dr Ilzetzki, to the specifics of the political economic problem we are trying to solve.
Regarding specifically the OBR, I had a little smile when I heard you because, quite frankly, in the community of independent fiscal institutions, the OBR is considered as a top-notch institution. I would not change anything to the OBR.
I want to react to one proposal made by my colleague who referred to scoring party platforms. We have that in Belgium. It was introduced about 10 years ago, so we have had two cycles now. Of course, Belgium is quite specific because we have many political parties. Some simply cannot produce things that can be scored. We have a political party that wants the independence of Flanders, and they say, “We are going to make Flanders independent”, so they say, “Do you want to score that for Belgium because it will not exist anymore?” Or the party platform is tax the rich. “Yes, but how will you do that?” and so on. It is difficult to do. I know very well the head of the Federal Planning Bureau, the equivalent of the OBR in Belgium, and he tells me it is a nightmare.
I would simply conclude: be careful what you wish for. It was clear in the latest scoring that we had before the latest election in Belgium, a bit more than a year ago, and guess who won the election? The party that won the election was the least fiscally responsible of all of them. It was the party that was promising tax cuts without any compensation and thinking that somehow expenditures will be adjusted later because deficits will be too high. Yes, but at the end of the day, this strategy—which has a name in the economic literature; it is called “Starve the beast”—does not work. We have seen that in the past. Countries say, “We cut taxes and then somehow deficits will be so ugly that we will have to cut spending”. That does not work.
So I suggest a touch of caution about giving the OBR the task of scoring political platforms. It is very resource intensive. If in the general public, there is not a shared understanding of what the government budget constraint means, or for that matter that there is a government budget constraint, you might end up having very weird results and the things you do not necessarily want. So I would say a word of caution about the scoring of electoral platforms.
Lord Razzall: But there is another word for it, which is democracy.
Dr Xavier Debrun: Yes, of course.
Dr Ethan Ilzetzki: If I may, I would express one more thought on this question but in the form of a question to my colleague, who has more expertise on fiscal rules across countries: Xavier, what examples can you give from other countries of allowing a medium-term target to really apply in the medium-term and have a more gradual adjustment? I think part of the problem that we have in the UK is this lack of medium-term planning that leads to this Budget-by-Budget change in policy? Could you give examples of other countries that have resolved that problem?
Dr Xavier Debrun: Yes, I would say the Rolls-Royce of fiscal frameworks with a medium-term perspective is Sweden. In Sweden, they have an explicit expenditure ceiling, which is anchored in a structural budget balance rule, and they define it over four years—or is it three years? It is three or four years. The outer year is not binding, but the other ones are binding. In each budget cycle, only the outer year gets revised. The expenditure ceiling, which is defined in nominal terms, is binding. The budget envelope is set for three years, and you cannot deviate from that. Of course, there are escape clauses in case of big shocks. However, if you lock in your nominal expenditure, you have a very stabilising fiscal policy.
Q40 Lord Burns: One issue that is causing tension around Westminster these days is the issue of dynamic scoring. I cannot believe that this has become an issue, and I am almost embarrassed to raise the question. There are two scoring issues that the OBR has. One is that it is asked to score tax and spending changes that the Government make. The second, of course, is that it makes forecasts twice a year. One question at the heart of this is how far it needs to consider second-round effects. When you do a forecast, you need to take into account any second-round effects, but there is an issue about, when you state what the cost of measures are, how far you do this. My view is that the people who argue for dynamic scoring are people who like to believe that there are longer-term magical effects of the measures that are being proposed. The question is: how far do you think the OBR is following what you might describe as best practice in its modelling work? How far is this an issue that other Governments face?
Dr Ethan Ilzetzki: Using dynamic scoring is the norm, I would say, and it makes sense to forecast the best we can. I do not think anybody has particularly good tools to do this dynamic scoring. There are a lot of assumptions—even more than the assumptions we usually make when it comes to scoring policy statically. When I look at what the OBR has done, it is doing the best it can with conventional tools but occasionally there are baked-in assumptions that can lead to things that are contrary to any empirical evidence.
The example to me was the first Budget of this Government, where an increase in public investment was forecast to decrease GDP within a five-year horizon. There are a wide range of estimates of what public investment does to the economy, but shrinking the economy is not one of them. It was using conventional modelling tools; it is not that what it was doing was entirely imaginary. But I think if your dynamic scoring leads to a result that is contrary to existing empirical evidence, you need to be cautious in altering policy. That was, in large part, what led to the increase in taxes in the first Budget of this Parliament. We would not want to be in a situation where specific modelling assumptions are leading to these results.
I was asked to get into the weeds, so I will just say one sentence on the weeds. In empirical work, we have much more knowledge on the causal effect of taxation than we do on public expenditure, just because public expenditure is so heterogeneous. There are five or six different types of taxes that Governments use, and we have empirical evidence on each one of those categories of taxes. Public expenditure includes education and the NHS; there are all kind of categories of spending, each of which has a slightly different multiplier. We collectively do not have the ability to forecast them dynamically as accurately as we can for changes in taxes. Whether we are increasing expenditure or cutting expenditure, I think it leads to dynamic forecasts that are far less precise than the dynamic forecasts achieved when forecasting the effects of tax changes, which themselves are imperfect but where we have much more to lean on.
Dr Xavier Debrun: Yes, on my side, I completely agree with what was said. On dynamic versus static scoring, dynamic scoring is really the state-of-the-art, simply because we tend to think that it would not be correct in any forecasting exercise to assume that there would not be any impact of fiscal measures on the economy itself. You must choose between two evils: one where you are sure to be very wrong, which is to do static scoring, and one where you try at least to be less wrong than you could be. Dynamic scoring is the norm.
We need to be careful also when we do that about the measures that we take. Let us not forget that models, however good or bad they are—but however good they are—and predictions based on models are always anchored in the hypothesis that somehow the economy in the future will behave like in the past. Many of the fiscal measures, not just expenditure but also specific tax measures—and the tax code is very complicated—can have deep implications for firms and households’ actual behaviours. The assumption behind many models is that those behaviours are constant over time. If you take measures that are known to have or that are even designed to have specific implications and effects on people and firms’ behaviours, you are more likely to get it wrong. As in any forecast, there is also a lot of judgment that must come in addition to just running the model and looking at the results.
Lord Burns: Can I relate this to the issue about scoring party manifestos? What impact does this process have on the reputation of the people who are doing the scoring in those countries that have this? I find it difficult to imagine that we would not be in the middle of an enormous political row with the institution if at the time of an election it were given the task of scoring each party’s manifestos.
Dr Ethan Ilzetzki: The good fortune of forecasters is that so much changes by the time the forecasts have to be evaluated that nobody really holds them accountable for them. Everybody understands that reality has changed by the time their forecast does not materialise.
Lord Burns: Does it affect their reputation?
Dr Ethan Ilzetzki: I have not seen cases. There may be in countries that I am less familiar with, but I do not think the media comes back and grills the Congressional Budget Office for mis-scoring a certain policy.
Dr Xavier Debrun: I fully agree with that. I think that, if there is a loss of reputation for forecasters, it usually occurs after a large, unusual event, like a crisis we could not predict. Then definitely, “How could you not predict that?” and you say, “Well, I do not have a crystal ball. I do not know. This never happened before”. I remember being in the middle of this situation when the pandemic started. Of course, Belgium, as a lot of times, did not have a Government at that time. We were at the central bank and elsewhere running different scenarios to try to inform the acting Ministers about what was going on, and we were working with Excel spreadsheets because not a single model could work. We were making crazy assumptions, although we ended up being not too wrong. You cannot be held accountable as a technician for that, but people did not blame the technicians. If you look at what happened in France, which is a situation I referred to before, the large forecast errors on government revenues were blamed on politicians and not on forecasters.
Q41 Baroness Liddell of Coatdyke: There has been a bit of a debate here about whether the OBR should produce only one forecast per year, and it would be interesting to hear from both of you about what you think is an optimal way of doing it. Should the OBR be responsible for its own macroeconomic forecast, or should it be the responsibility of the Treasury, and how do other countries do it?
Dr Ethan Ilzetzki: I think the norm is to have one forecast a year. In that respect, the tradition we have maintained here of two a year is unusual. With that in mind, I do not think it would matter as much if, again, these forecasts were not so binding on government behaviour. In that case, more information would be better and there would be nothing wrong with having two forecasts a year.
Q42 The Chair: Before Dr Debrun comes in, can I just ask you: I thought the IMF recommended that countries should have two forecasts a year. Is that right?
Dr Xavier Debrun: It is not so unusual to have two forecasts a year. In Belgium, we have two forecasts a year. There is the forecast that is used for the preparation of the Budget; there is a second forecast that comes in February of the Budget year. If the forecast is really off, the Government convene and they hold what we call a conclave, and they decide on supplementary measures if the targets are not attained. Again, this might be too much high frequency, too much obsession with specific numbers of a fiscal rule, but that is a norm.
For the rest, yes, there are many countries I can think of that have two rounds of official forecast. What matters is not so much how many forecasts a year you produce, because it could happen that exceptional events again occur and force you to revise your forecast, if only to look credible as someone who knows what is happening right now and what are the implications of what is happening for the future, so there is your reputation purely as a forecaster.
What really matters at the end of the day is what do you do with your forecast. If you were to have four forecasts a year because you consider that every quarterly number for growth is bringing sufficiently fresh and new information to change your forecast for the next three years, okay. The main thing is that it does not lead to changes in fiscal policy four times a year because you have a new forecast. It depends on what you make of this forecast.
The Chair: Great, thank you very much. We have done a comprehensive tour, not just of the UK’s fiscal architecture but of many other countries as well. We are grateful to both of you, Dr Debrun and Professor Ilzetzki. Thank you so much. We have learned a lot and appreciate your time. With that, I can say that the meeting is now concluded.
