Keeping Government responsible
Robert Chote, chairman of the Office for Budget Responsibility, speaks to the Journal about the OBR's work, ahead of a meeting with the leaders of Scotland's big firms
What brings you to Edinburgh, and what is your message to business at events such as this?
I come to Edinburgh once or twice a year. One of the things I do is give evidence to the Finance Committee of the Scottish Parliament, who are generally interested in the outlook for the economy and the public finances. We've been given responsibility under the Scotland Act for forecasting the set of taxes that are due to be devolved to the Scottish Government, including the new land and buildings transaction tax.
Another interest is to meet with colleagues from the Scottish Fiscal Commission, who are in a sense our new opposite numbers here. Their role is potentially going to be increasing as a consequence of the Smith Commission and we need to make sure that we have a good working relationship with them, and with the Scottish Government: we will continue to have to produce forecasts for the things that are devolved to Scotland, because we have to produce a set of fiscal forecasts for the UK as a whole, including the devolved administrations, public corporations, local authorities etc.
In terms of talking to businesses, which I often do while I'm here, it's helping them to understand what we are doing as an institution and our latest views on the outlook for the economy and the public finances. We're in a slightly weird position at the moment as we've got a general election looming, and the set of forecasts we produced at the Budget are not a description of what either member party of the coalition would wish to do on their own, but they are the baseline that whoever becomes the new UK Government will have to start with and explain what they want to do differently and how.
I would have thought you would all go on holiday during an election campaign, because you are not allowed to comment on individual parties' proposals!
You would like to think that; unfortunately it's not how the weeks running up to it are working out! For one thing we have a selection of publications that we generally produce over the summer – long term fiscal projections looking 50 years ahead instead of five years, where we bring in things like demographic change or the evolution of student loan systems; we also produce an annual publication that looks at the performance of our past forecasts and explains, if things turned out differently from what we thought they might, why this was; and one we have more recently been asked to produce, on drivers of welfare spending. All these we would normally do over the summer, but we are trying to concertina as much as we can into the next few weeks, because our working assumption is that whatever the election result, we will have an extra Budget to cope with over the summer as a new Government comes in.
Is this work intended to inform the election debate?
Not strictly speaking. We produced a forecast at the time of the Budget, which was based on the policies that the coalition managed to agree on public spending over the next five years. But each party simultaneously said that if they were to govern alone, they would do things differently, so it's an odd situation because we are required to produce our forecast on the basis of current Government policy. So we have in that sense provided some background to the election; what we don't have to do is look at the implication of what the parties are saying in manifestos. That was raised as a possibility a few months ago, but Parliament decided it didn't want us to do that. In my previous job at the Institute for Fiscal Studies [IFS] I did do that, ahead of the previous two elections, so I'm now cheerfully looking at my successor having to shoulder that burden over the next few weeks.
Is your remit too narrow if you can't comment on policy issues or engage in that kind of analysis?
I think there is value in having a relatively tightly defined remit. We have the IFS to do that sort of thing. You could have a more formalised arrangement for an organisation like ours to look at the proposals. The most fully developed is in the Netherlands, where our Dutch counterparts formally scrutinise the manifestos submitted to them by the political parties well ahead of an election and produce a lengthy report. I think in the last election they had to do it for nine parties and 2,500 policy commitments; it took 80 people three months. That would be quite a mammoth exercise for us, even with a smaller number of parties.
This was debated when we were set up; there was I think cross-party agreement at the time that it would not be sensible for the OBR to do it until it had established its reputation through one political cycle at least. At the time there was huge controversy over how to repair the public finances, and throwing a newly born organisation right into the middle of that very politically contested territory would have been very dangerous. Labour then proposed a couple of years ago that we should be willing to look at the costing of individual policies put forward by opposition parties, but that didn't get cross-party support.
I think there's a case in the longer term that having proper independent scrutiny of party platforms is good for democracy; the Dutch experience also suggests it's helpful when you have to form coalitions and meld policies together. However it does have some interesting logistical consequences – we would need a lot more people, although only for the latter part of the political cycle. It also raises some interesting constitutional questions. Unlike our Dutch equivalents, we're not a big office, but we have a legal entitlement to the time and effort of people in HMRC, Department of Work & Pensions etc, and in a sense it would raise the issue of whether civil servants should be providing support for opposition parties developing policies, and that I think is something that, reasonably, people want to think carefully about, rather than simply accepting it as a backdoor consequence of giving us a greater role. But it's something that the Treasury Select Committee is very keen to look at again once this election is out of the way.
Would you have an advantage over the IFS, with more access to civil service analysts and a better understanding of what is going on?
Yes. When I was asked how we would go about this, I basically said that we should do it in the same way, as closely as possible, that we scrutinise Government policies now in Budgets and Autumn Statements. You would want to do it with the same degree of detail and authority and transparency. That takes quite a lot of time and effort: what you don't want to do is give somebody a back of the envelope calculation; they then win the election, implement the policy and we say actually this is going to cost twice as much as we told you six weeks ago. That would be bad for them and bad for us.
The other big adjustment, if the OBR were to be given power to assess political parties' election promises, would be a huge revolution in the parties' behaviour. If we were to have this role we would have had to have been told months ago about the policies that the parties wanted to announce, in order to go through them properly, and parties would need to finalise their manifestos in private much earlier than is currently the practice in Britain. That's certainly the Dutch experience, so I think it would be at least as big a leap for the political parties as it would be for us or the civil service.
Have you actually pulled ministers up for overoptimism in the course of your operations?
It's not so much pulling them up. What's different in this arrangement from the way it's done in most other countries is that the Government has basically outsourced the task of producing the numbers to us. So the Government doesn't give its view and we come up with an alternative; we come up with a view and the Government has to decide whether it will adopt that forecast as the basis on which it will set policy, or – which hasn't happened yet – it wishes to set out an alternative, which would be entirely within its right. It's not just symbolic: the burden of proof is in a different place. We're basically saying that we produce the numbers and the politicians react, and not that the politicians produce the numbers and we react.
Is it possible to completely avoid commenting on policy in the course of your analysis?
Not entirely. We are precluded in our legislation from reaching a view on whether something is a sensible policy or a stupid policy; however we do have to produce an estimate of how much it will cost or how much it will raise, and in order to do that you have to explain what impact that policy is likely to have. So if you have a policy that is designed to achieve a particular objective and our costing is based on our judgment that it's not going to achieve that objective, people can draw their own conclusions. But it has to come out of our assessment of the costing. It's not our job to say that only an idiot would do such-and-such.
Do you have a view on how big a budget deficit is sustainable in the longer term?
There's no hard and fast rule on that. There are a number of things that determine what sort of deficit you can sustain over the long term while keeping your debt on a sustainable path. A lot depends on the relationship between economic growth and interest rates. The higher the interest rate, the more your debt builds up; the higher the growth rate, the bigger the flow of revenue there is to service it with. What we do is basically to produce a projection with a five year time horizon at the time of the Budget and the Autumn Statement, based on existing policy. We judge that looking at some of the headline indicators of the public finances, most obviously public sector net borrowing, which is the budget deficit, but the Government also has sets of specific targets, and one of the things we have to do is to say, not whether it's the right target or the wrong target, but are they on course to hit it. They have an objective to get the public sector deficit falling as a share of GDP by 2016-17; we think they are on course to achieve that. They have also in the latest Budget got themselves on course to achieve it a year earlier, but they are doing that by selling assets, not by increasing revenues, as we pointed out.
Is it realistic to think in terms of finances moving into surplus in the next Parliament? Would it involve very radical changes to current spending patterns?
Crucially, if you look over the next five years and take the coalition plans that are written into the Budget at face value, the key thing looking forward is that about 70% of the hard lifting that is being pencilled in in terms of reducing the Budget deficit and getting us into surplus involves cutting day to day spending on public services. We're not getting an enormous amount more tax revenue; we're not cutting capital spending by very much more; we're not cutting welfare spending by very much more. Now some people from outside would look at that and say it just can't be achieved. Famously, back in December we pointed out that this involved public spending shrinking to its lowest level since the 1930s. The Government has made sure that by a small margin that is no longer the case by adjusting policy in the Budget at an aggregate level, but I don't think it is appropriate for us to make an ex ante judgment that you cannot shrink public spending on public services to that sort of level. That's ultimately a political choice, at least as much if not more so than an economic one.
What we have to do is look in real time at whether the Government is sticking to the spending plans it set out for itself; if it were failing to stick to its plans for the current year, we would have to look more sceptically at the claim or assumption that it would do what it said over a five year period. But it wouldn't be appropriate for us to say, ex ante, this isn't doable – the obvious response to that would be, OK, what do you think is the level of public spending to which the Government can cut? And as I say that's a political choice at least as much as an economic one.
If the official inflation rate went below zero, would that be a concern to you – does it make targets harder to achieve?
Not necessarily. The different measures of inflation affect the fiscal outlook in different ways. For example, movements in retail price inflation affect how expensive it is to service index linked gilts, how excise taxes move, how much they are raised by if governments don't do anything about it – the default changes. Movements in consumer prices affect increases in tax allowances, tax thresholds and benefit levels.
There is a broader concern people have about whether you end up in sustained deflation: when people think prices are going to fall year in, year out, that creates a powerful incentive for them not to spend money today because things are gong to be cheaper tomorrow, and that becomes a self fulfilling prophecy because it weakens things further today. The Bank of England is clearly very alert to that danger and has a target to get inflation to 2%, and we presume that the monetary policy it is following is consistent with that, so in the forecast we produced at the time of the Budget, we have inflation at near zero as a quarterly average until the middle of this year, and that is quite consistent with consumer price inflation being negative in some months, but I think the consensus would be that this is a relatively benign form of deflation because oil prices have come down so far – that basically means that the take home value of people's incomes has increased, and that's positive, rather than being a Japanese style deflation in which people pull in their horns and that just makes things worse.
How much impact would interest rates have on your projections if they start going up again?
We presume that they will go up over the next five years. We basically take the interest rate projections that are built into financial market prices, so we don't forecast interest rates ourselves – we assume they are going to move in line with market expectations. If they were to rise more quickly than is currently anticipated, the fiscal impact would depend crucially on the reasons why. If it is because the economy is doing better than expected and people's incomes are growing more rapidly, that's not necessarily a bad thing for the fiscal position because tax revenues would be more buoyant, or for people's personal finances because while their mortgage bills will be going up, so too will be the incomes they have to pay them with.
What we would be more worried about would be if interest rates went up more quickly than expected but incomes and earnings didn't, for example because of conditions in the eurozone, greater anxiety about bank funding costs etc, basically interest rates rising because people are worried about risk. If you have the burden of interest rates rising while people's incomes aren't, that creates the potential for difficulty, and if it's associated with weaker potential economic performance in the future, that's not good for the economy or for the public finances.
Are there any other relevant issues that we haven't covered?
Perhaps one issue from our point of view is that we look forward with great interest to seeing what finally emerges out of the Smith Commission process. As I say, we will continue to have a responsibility to the people of Scotland as much as to the people of the rest of the UK to produce an assessment of the public finances across the entire public sector in the UK, but clearly it's going to be a more complicated process if more decisions are being taken in different places in the country, not just Scotland but Wales and Northern Ireland as well. I think it will be very important to establish a good working relationship – and it is going very well at the moment – with the Scottish Fiscal Commission, with the officials of the Scottish Government, with Revenue Scotland as it builds up, because we all have a shared interest in being able to predict these things as best we can. The Scottish Government needs to produce a set of forecasts, and it's independent for that purpose; it doesn't matter if our forecasts are not the same, but we at least need to be able to explain why they are different if they are different. That's an important issue.