Treasury Committee
Oral evidence: Economics of HS2, HC 788
Tuesday 5 November 2013
Ordered by the House of Commons to be published on 5 November 2013.
Written evidence from witnesses:
Members present: Mr Andrew Tyrie (Chair), Mark Garnier, Stewart Hosie, Andrea Leadsom, Mr George Mudie, Mr Brooks Newmark, Jesse Norman, Mr David Ruffley, John Thurso
Questions 1-144
Witnesses: Richard Threlfall, Partner and Head of Infrastructure, Building and Construction, KPMG LLP, and Lewis Atter, Partner, Global Infrastructure and Projects Group, KPMG LLP, gave evidence.
Q1 Chair: We are going to run this session to time if we possibly can. This is a very big subject, and I expect we will be wanting to look at it again, and I hope you will understand that we may need to ask you to come back at some later stage for further information about this very important subject. This is a huge sum of money by any standard. It is one of the biggest public spending decisions ever undertaken by a government.
Can I begin by asking you—I do not know to whom I should address this question—how much did you charge for this?
Richard Threlfall: Good morning. Thank you, Mr Chairman. In terms of the questioning, perhaps it might help if I explain that I am Head of Infrastructure, Building and Construction at KPMG and I am ultimately responsible for any work that KPMG undertakes for High Speed 2. My colleague, Mr Atter.
Lewis Atter: I head up our infrastructure strategy business, which seeks to connect economics, funding and financing and decision making.
Richard Threlfall: In terms of your question, Mr Chairman, the fee for the work was just under £200,000. Including VAT on top, I think it has been published at £242,000.
Q2 Chair: The full cost is £242,000?
Richard Threlfall: Including the VAT. That is correct.
Q3 Chair: There are no further costs—
Richard Threlfall: For this piece of work there were no further costs, Mr Chairman. That is correct.
Chair: —that are likely to fall to the Government? Thank you. I would like to turn to the measure of productivity that is being used. Take the £15 billion figure that is in circulation, that is the £15 billion per annum benefit for the life of the project, which I think is 60 years, calculated on a 60-year basis. Is that a considered estimate, or is it something less than that?
Lewis Atter: I would say it is a considered estimate. We have based our approach here on the best available evidence. We have asked the questions about what we can see in the real world in terms of differences in economic performance between cities, how much of that is explained by differences in connectivity, based on real world examples out there today, and we have asked what difference HS2 will make to that, and, therefore, what difference HS2 can make to the productivity of the UK economy. It is very considered, based on the best available evidence today.
Q4 Chair: It is a reasonable forecast of what we might hope to get from this project?
Lewis Atter: We believe so.
Q5 Chair: The reason I use the word “forecast” is that I note that in several places you caution that a key part of your analysis is, and I quote, “Prepared for illustrative purposes only and does not constitute a forecast”, and a moment ago you told me that this considered estimate is a forecast.
Lewis Atter: It is a considered estimate.
Chair: But it is no longer a forecast?
Lewis Atter: I think it is clearly a forecast.
Chair: Even though you are making clear that it does not constitute a forecast in your own report.
Lewis Atter: No, I think you are looking at our standard disclaimer.
Q6 Chair: This is a standard disclaimer?
Lewis Atter: Yes.
Chair: So if you are forecasting that tomorrow the sun will rise, you will nonetheless put a disclaimer on and say that is only a projection?
Lewis Atter: I think the conversation with Risk and Compliance might be a little more straightforward, in that case, but yes.
Q7 Chair: This is to stop you being sued?
Lewis Atter: Correct.
Q8 Chair: You would not want to say, therefore, that this is a forecast outside this room, and you are saying this because you are under parliamentary privilege. Is that it?
Lewis Atter: I think in the sense of this discussion it is a forecast.
Q9 Chair: Let us have a look at this £15 billion figure. Can I take you to page 84 of your report? What you have done here, as I understand it, is quite a complex calculation. You have tried to estimate the relationship between productivity and connectivity using four measures. The problem with all of these measures is you want to know whether the numbers you are getting are reasonably reliable and not just a rubbish in, rubbish out computer model. Correct?
Lewis Atter: Correct.
Q10 Chair: You would have run a diagnostic test on all this?
Lewis Atter: We have run a number of tests on this. Perhaps if I step back a bit—
Chair: By all means step back in just a moment, but I just want to clarify a few points about these numbers. You ran a diagnostic test or series of tests on this.
Lewis Atter: Series of tests, yes.
Q11 Chair: What did those tests tell you?
Lewis Atter: We ran a series of tests. Those tests told us, not surprisingly, that one of the things you need to worry about when you are looking at relationships like this is whether the one you are particularly interested in—in this case, the most important here is the impact that improving business connectivity by rail has on productivity. What you really need to be worried about in that context, particularly given the question of what impact HS2 will have, is whether what you are seeing out there in the real world data in terms of differences between business-to-business connectivity and differences between productivity is solely a function of differences in business-to-business connectivity by rail, or partially a function of labour connectivity by rail, whether it is partially a function of road connectivity to businesses, or road connectivity to labour. Against that background, and given the exam question we were posed, we said the most important elasticity in all of this is that between differences in business-to-business connectivity by rail and productivity. What we need to worry about is whether that elasticity is partly to do with other things.
We took that elasticity and we spread it across the other things. We said, “Yes, this is our most important elasticity, but there is a risk that what we are seeing here is a function of differences in labour markets. It is a function of what is happening on the road network both in terms of business connectivity and labour connectivity”. We spread that elasticity across the other impacts. That had the result of—transparently set out in our report—more than halving the responsiveness in—
Q12 Chair: This is a crucial piece of analysis lying behind the £15 billion figure, is it not?
Lewis Atter: It is a very important part of it, indeed.
Chair: It is the crucial, most important single building block of the productivity measures and the relationship between productivity and connectivity. The coefficients for the connectivity will determine what productivity—
Lewis Atter: I would agree. It is the single most important number in there.
Q13 Chair: All right. You have done those diagnostic tests, and what did they tell you?
Lewis Atter: They said that we needed an approach something like this one. We also discussed the approach with the advisory panel, and they advised perhaps we should also look at alternatives.
Q14 Chair: Sorry, just for everybody here, the advisory panel?
Lewis Atter: HS2 appointed an expert advisory panel to assist.
Q15 Chair: You went back to the advisory panel for HS2 to seek advice—
Lewis Atter: We had a number of conversations.
Chair: —on how to run your independent piece of analysis?
Lewis Atter: The panel is made up of independent experts. We had a number of conversations with the panel about this approach and alternatives. Some of the panel thought we should also address the statistical issues here by blending our approach between measures of connectivity based on rail, measures of connectivity based on road, by forecasting a market share between them. All right. You will find in our report we also show the impact of that approach, recognising that, in using that approach—it is one of the challenges in this—you have to forecast road and rail market share.
Q16 Chair: You have told us before, a moment ago, that this is a forecast, you have confidence in it, it is robust, and, just now, that that figure is very heavily based on this piece of statistical work. Do you feel that piece of statistical work can be relied on?
Lewis Atter: I think in the circumstances, and given the exam question we asked, it is a reasonable approach. I would point out—
Q17 Chair: Do you think it is reliable?
Lewis Atter: If anything, it feels a little undercooked. I will point out why.
Q18 Chair: What do you mean by undercooked?
Lewis Atter: It looks a little low. I will point out why.
Q19 Chair: In other words, it is unreliable because it is too cautious?
Lewis Atter: I think it is conservative, yes. One of the reasons—
Q20 Chair: I am trying to just get a very simple answer to a simple question, which is, do you think that we should be able to rely on this part of your statistical analysis for the purposes of assessing this project?
Lewis Atter: We are satisfied, given the questions that we faced and the nature of the data and the state of the art as we find it, that it is a reasonable approach.
Q21 Chair: Is it a reliable approach?
Lewis Atter: We think it is.
Q22 Chair: You think it is a reliable approach?
Lewis Atter: That is right. If anything, it is on the low side.
Q23 Chair: I have understood that you are saying that if it is unreliable it is because it has been biased in the conservative direction. If anything, the middle point might be higher than this figure.
Lewis Atter: We thought it was only right, Mr Chairman, in terms of transparency that we set out uncertainties both positively and negatively.
Q24 Chair: We are going to come on to that in a moment in later questions. Can I take you to page 83 of your report, paragraph 6.3.38? You have just told me that this approach is reliable and it is crucial to the analysis of the whole of this report, but you also say, and I am quoting, “We recognise that this approach does not have a firm statistical foundation”.
Lewis Atter: Indeed.
Chair: This is an approach that you think is reliable, but it does not have a firm statistical foundation. This is quite a tricky piece of territory to sit on, is it not?
Lewis Atter: Against the background of the question we have been asked, the data that is available and the state of the art, we think all of this is reasonable. We have been very transparent about the adjustments we have made to recognise the uncertainties here. One of the standard approaches—I talked about this a few moments ago—in the face of these challenges is to use a blended measure of connectivity, so we will mix and measure road connectivity with a measure of rail connectivity. In order to undertake that mix, you have to forecast market share. You have to forecast what road use is and what rail use is right the way across the network in a future year, and that future year is in the late 2030s. You can, if you want, substitute this sort of approach for another one. The other one requires you to forecast market share with some accuracy in the late 2030s. There is no such thing—
Q25 Chair: I will just try to clarify where we have got to this morning. You have agreed that the figure of £15 billion per annum return for 60 years is a forecast and not just a projection.
Lewis Atter: It is a forecast of an annual impact in the late 2030s. It might vary over the—
Chair: I am only summarising what you have just said. You have agreed that that is heavily based on a piece of statistical analysis, which you initially told me was reliable, and your description of it as reliable you feel is consistent with what appears to be another disclaimer in your report, which is that this approach does not have a firm statistical foundation. To a layman, and not just to a layman, that does seem something of a contradiction, Mr Atter.
Lewis Atter: It is the nature of this particular beast. We have been very transparent about the adjustments that we have made to the elasticities that drop out of the analysis in light of the risk that the number you are seeing here is in practice a reflection of other differences in connectivity.
Can I also point out a couple of other important areas of conservatism in this analysis? This analysis and these elasticities look at all places on an even basis, with essentially an average relationship, right the way across all the cities in the UK. All right. In practice, we strongly suspect, on the basis of analysis we have been doing subsequently, that the elasticities—
Chair: That is not published here?
Lewis Atter: Not published here, but the elasticities you see in respect to changes in connectivity for the most connected places—in practice the places that would be impacted by high-speed rail—are significantly higher than this. Point one.
Point two, in terms of conservatism, this analysis, as I said right at the beginning, is based on relationships between differences in connectivity and economic performance today. All right?
Chair: Perhaps it would be helpful if you could set out in writing for us after this meeting the first point in as much detail as you can.
Lewis Atter: Happy to do so.
Q26 Chair: That would be very valuable because you are supplying information that has not been published and on the basis of which we cannot possibly comment, and if you want to add other points that are in the report—I think the second point is in the report—that would be helpful.
Could I just take you back to that sentence on page 83 before I pass the questioning on? It reads, “It enables connectivity to other business and to labour, by car and rail, to be reflected in the analysis”. There appears to be a contradiction between the first half of this sentence and the second half of this sentence for estimating what the benefits will be for most of the rest of this century. What is the meaning of the last two words of that sentence?
Lewis Atter: I think there is a typo. “To be reflected in the analysis and captured.” The point here, if we go back to the—
Q27 Chair: Was there a bit more of this sentence there, do you think? Has it been cut?
Lewis Atter: It is, “the analysis and captured”, rather than “captures”. I think that is a typo. Can I just—
Chair: Very briefly, but a lot of colleagues want to come in.
Lewis Atter: Absolutely. If we go back to what we did in respect to this key elasticity, we took the key elasticity and divided it among the other measures of connectivity. What that allows you to do is look at the bit of elasticity you took from business connectivity by rail, and we put it over here in respect to the road, in terms of labour, in terms of road, in terms of business, and rail in terms of labour. You have that spread of elasticity, sourced from one, and spread across the others. Then you use those elasticities as and when something like HS changes labour connectivity by rail or connectivity by road. That is what that sentence means.
Chair: That is very helpful.
Q28 Mr Ruffley: I want to ask about the regional economic impacts. Just to explain what I am most interested in, I just want to re-tell what Professor Henry Overman of the LSE, formerly an expert adviser to HS2 Limited, has said. “When a firm is thinking of where to locate, it thinks about the relative productivity of different places and the relative wages. HS2 shifts that around.” So, if you are on the line that makes you a better place but have not had that productivity improvement, you will be in a better place. That much is obvious; there will be winners and losers. Why were the figures on the cities that potentially could lose out as a result of HS2 omitted from your final report?
Lewis Atter: They are not omitted from the final report.
Q29 Mr Ruffley: I am thinking of the granular stuff that Newsnight extracted under an FOI request. Why wasn’t that in the final version?
Lewis Atter: If you look at page 54, there are two maps that address the distribution of economic output that we are forecasting as a result of HS2. Very clearly, these maps have positive values in green and negative values in red.
Mr Ruffley: Hang on—
Lewis Atter: We took the view that this was a very transparent and clear way of making the point that there were places that would win as well as places that would lose, and showing people where we thought those differences would be.
Q30 Mr Ruffley: Just so I am clear about this, the FOI request that Newsnight have produced, that is not in the report, is it? I am asking you, why not?
Lewis Atter: We thought this was the most appropriate way to explain this disaggregation, and in terms of the overall impact, this is—
Q31 Mr Ruffley: Why more appropriate? You had this information and it lists 235 areas. I am referring to the FOI request that was not published. This is information you had and did not disclose willingly. To take my own area, Suffolk West, as an example, it argues there is a minus 1.4% fall in GDP, and that is the estimated change in output, measured in terms of the potential impact of investment in HS2 on annual economic output for each area. There is quite a granular set of losers here, and I just repeat the question. You gave us a map. Why didn’t you produce the information that you had to hand, in the interests of openness and transparency?
Richard Threlfall: If I may, Mr Ruffley, we took a view that it was clearer to set it out in the form of a map so that people could see visually where the impacts would be positively and negatively, rather than to publish a table of numbers that mix those up, the 254 areas. We supplied all of that data to HS2 so that if people asked for it, as they clearly did, it would be available. If somebody had asked us for the numbers, we would of course have been happy to provide them.
Q32 Mr Ruffley: All right, Mr Threlfall. Was there pressure from HS2 Limited to leave out this much more granular information?
Richard Threlfall: No, absolutely not.
Q33 Mr Ruffley: Whose decision was it to leave out this granular information that the BBC had to ask for an FOI request in relation to?
Richard Threlfall: It was not a decision to “leave out” the data.
Q34 Mr Ruffley: Well, you did leave it out.
Richard Threlfall: It was a deliberate decision that we thought it was a lot clearer to show the position on a map with red blobs for the negative impacts.
Q35 Mr Ruffley: Mr Threlfall, whose decision was it? Who actually took the decision?
Richard Threlfall: KPMG. We took the decision.
Q36 Mr Ruffley: Who in particular?
Lewis Atter: I think I remember discussing it with the team. We were looking for—
Q37 Mr Ruffley: It was your decision, Mr Atter?
Lewis Atter: Absolutely. Absolutely.
Q38 Mr Ruffley: It was your decision. All right. The TaxPayers’ Alliance said it was, and I quote, “Utterly dishonest to have spent £250,000 of taxpayers’ money on a report and then to release only cherry-picked findings.” Mr Threlfall, what would you say to that charge from the TaxPayers’ Alliance?
Richard Threlfall: I would say we would completely reject that charge. We think we have been—
Mr Ruffley: You did not cherry-pick?
Richard Threlfall: We did not cherry-pick the results.
Mr Ruffley: You did not cherry-pick?
Richard Threlfall: We think we have gone out of our way in this report to be as transparent as we could possibly be. We have been transparent about the methodology, even returning to the prior exchanges about recognising some of the limitations in the methodology, some of the upsides and the downsides. We have been very transparent about the assumptions and about the outputs of the report and captured in this map.
Q39 Mr Ruffley: Your report states, and I quote, “The analysis is based on data and assumptions used to suppose the August 2012 iteration of the economic case.” Can I ask, were you asked to update your analysis to reflect the changes and assumptions made for the October economic case?
Richard Threlfall: This report was clearly published in September, so it pre-dates the—
Q40 Mr Ruffley: I am asking you, were you asked to update?
Richard Threlfall: Not formally, no. We are discussing whether that would be a sensible thing to do. It is clearly a matter for HS2 if they now ask us to update this report in the light of the new assumptions.
Q41 Mr Ruffley: So you are waiting for HS2 to give you an instruction?
Lewis Atter: Also for the data. We do not have the data.
Q42 Mr Ruffley: Also for the data. That is helpful. How significant an impact do you think the changes in the data and assumptions for the October 2013 iteration would have on your analysis? We are just looking at a general feeling. Mr Threlfall. Sorry, you were conferring.
Lewis Atter: Yes. It is a bit speculative at this time. We see some areas—
Q43 Mr Ruffley: I am asking you not to be speculative, but to make an educated guess. You are both very experienced analysts.
Lewis Atter: We understand, but we have not seen the details. That—
Q44 Mr Ruffley: What are your understandings?
Lewis Atter: We understand that the new analysis of routes, services and the rest of it has additional effort into use of freed-up capacity. If efficient use is being made of freed-up capacity, we would expect overall productivity impacts to increase. We also know that whereas the 2012 version did not have the station at Manchester Airport, the 2013 one does. Both those things would suggest that the economic benefits would increase, but there may be things in the service pattern we do not know about that would reduce the number.
Q45 Mr Ruffley: Your provisional view, without having seen all the data because it has not been supplied to you in detail—is it is upside? No downside?
Lewis Atter: There is upside in the train plan, but there are other things.
Mr Ruffley: What are the other things?
Lewis Atter: You will notice that the Department for Transport has revised its value of time for business use. It has decreased it by about a third. We ran a sensitivity on this analysis, reducing it by half. When we reduced it by half, the net economic gain per annum was £12 billion, not £15 billion. The Department for Transport has reduced it by a third. We would expect that impact in isolation to have reduced the overall number, but there are other things that would push the number up again, because when this initial analysis was undertaken we had not looked at labour participation. We had not looked at the international dimension. We had not looked at freight. Again, those things could put it up. The list of things that would put the number up is longer than the ones we know about that would reduce it.
Q46 Mr Ruffley: That is useful. I am just wondering when this further work might be done. It sounds to me as if you are just answering questions to order, that you are answering the questions that HS2 put to you or the work they ask you to do. Do you not think it might be a good idea to suggest to them that this might need updating along the lines you have just rather helpfully set out?
Lewis Atter: We have indeed done that, and there is a section of the report that discusses areas for further work.
Q47 Mr Ruffley: When do you think that work might be done?
Lewis Atter: We are discussing that with HS2 at the moment.
Mr Ruffley: But roughly how quickly, do you think?
Lewis Atter: There are a number of things that need to be done. It would take a matter of months, roughly.
Q48 Mr Ruffley: I just have one final question.
Chair: Very quick, and a very quick reply, please.
Mr Ruffley: Your report says that it is based upon publicly available information provided to KPMG LLP by HS2 Limited. My simple question is, did you seek to challenge any of the information on costs and benefits provided to you by HS2 at all if you thought any of the information might be a bit iffy or unclear or nebulous? Did you query—
Lewis Atter: The cost of constructing HS2 has nothing to do with this analysis. In terms of the data we were provided, we asked deeper questions in detail in clarification of it, and as I explained earlier, we did some sensitivity tests. We did a test, for example, that halved the standard value of time for business uses on rail.
Mr Ruffley: Thank you.
Q49 Stewart Hosie: Thank you very much indeed. Can I go back to the Chairman’s initial question about whether economic benefits of this would be £15 billion a year by 2037? Why did you not discount the figures to give a present value?
Lewis Atter: To discount the figures to a present value we would have to forecast not just 2037 but every year from 2036 out to, gosh, 60 years or so beyond 2037.
Q50 Stewart Hosie: So you felt this was the most accurate figure you could come up with?
Lewis Atter: The issues that go to what the impact on the real economy is of High Speed really boil down to, what do you think the impact would be when the system is mature? Everything else is variations on that theme. Does the £15 billion go up over time? Is it flat? Does it go down? Key question, is it £15 billion in the first place? To address that, we focused on a snapshot year, a reasonable number of years after the system had opened.
Q51 Stewart Hosie: I will set aside the fact that the whole analysis appears to be outwith the standard Green Book approach, so it is difficult to compare it with what otherwise might have been concluded. However, you said earlier the estimates were undercooked. They were conservative, they were on the low side, if anything. In a sense it is good to hear you say that. However, the experts in the field like Professor Dan Graham at Imperial College have said the £15 billion figure, “Seems to be on the optimistic side”, which rather contradicts. Why is an academic saying that appears to be on the high side, while you are suggesting these figures are on the low side?
Lewis Atter: You will have to put that question to them.
Stewart Hosie: Well, I am putting it to you.
Lewis Atter: Some of the questions that have been asked about this analysis go to labour market impacts and quality of labour. That is a particular interest and area of expertise of Henry Overman, for example. His analysis, as in the past, suggested that a lot of the labour market impacts that you see on the back of investment are not a function of the size of an effective labour market but its quality. Given that, he has posed questions about whether these numbers are overcooked. In response, actually labour market impacts are a very small part of our overall number. It is less than £1 billion of the £15 billion. But there are a number of different perspectives on aspects of this analysis that people bring to bear, and it leads them to take different views about it. I think that is inevitable.
Q52 Stewart Hosie: That is helpful. Can you tell the Committee, is your method of assessing the regional impact benefits entirely new, or have you drawn on the work of others in working that out in the calculations?
Lewis Atter: We have looked at work of others. We have also had lengthy conversations with HS2 and the expert panel, and we sat back and asked the fundamental question here. It is part of the rationale for the way we did the work, but—
Q53 Stewart Hosie: As you said earlier, the rationale you used was basically averages across the economy, without looking at the impact in specific areas. Is that correct?
Lewis Atter: No. The responsiveness, the elasticity between changes in connectivity and productivity, that is one that this analysis averages. In terms of the impact on individual places, what is driving that is a combination of things. One is, how much more productive does HS2 make one place versus another place?
The other thing that is driving that is what we have done through HS to lower the barriers to trade and competition between them. Not only does HS make individual places more competitive and by different amounts, it also lowers the barriers to competition between them by lowering the costs of travel between them. That is why our analysis started with productivity and said, “Right, step one, what difference have we made to productivity across the country? Step two, how does that changed national cake on the back of those productivity gains now divide up differently?”
It divides up differently on the basis of what we have done to the ability of one place to compete with another for business, and it is that second bit, where essentially we built a model, a number of versions of it, that says, “Right, what do we think drives competition between cities?” A function of what the labour costs are, what their productivity is, but also what the costs of trade between them are. We built the model, a number of versions of it that ask that question. We calibrated that model to today, and then asked the question, given those relationships, what difference will HS2 make?
Q54 Stewart Hosie: Which part of that model and which part of your analysis are you least confident about?
Lewis Atter: That analysis is challenging because what it is really asking about is trade between places, and there are no trade statistics between places.
Q55 Stewart Hosie: You say it is challenging. If I go back to the granular figures that David Ruffley had, which showed Aberdeen and Dundee potentially losing out on a low business case scenario, taking into consideration everything you have said about elasticity and about trade between cities, for Dundee and Angus and Aberdeen to lose out in the way the low business scenario implied would require trade from Aberdeen, serving the North Sea, for example, with a fixed base, to be served from somewhere near Birmingham because it had a rail line to London. It would require people working in the bio life sciences industry at Dundee University to be served from somewhere near Leeds because it had a fast railway line to London. That does not really seem to add up, to me.
Lewis Atter: That is, of course, the Scottish economy today. All right? This is asking questions about the UK economy in 2037. It is a different place. We acknowledge that one of the things on our to do list is the international dimension to this. One of the things that is very clearly unique about Aberdeen is North Sea oil. That is the national-external dimension to this, and it is on our—as we identified earlier—to do list in terms of the further work.
Q56 Stewart Hosie: When you say identified a requirement to do further work, I get the impression you are not dreadfully convinced by some of this analysis.
Lewis Atter: At the overall level, at the regional level, yes. In terms of individual places, where there is a significant international dimension, there would be a question mark.
Q57 John Thurso: Just a very quick question, if you can help me. To what extent in your report do you take account of modal share and the potential pluses or minuses that may come from that?
Lewis Atter: In our central analysis we do not rely on forecasts of mode-share, but what we do have in our central analysis is the forecast impact of HS2 on road congestion. There is an impact as a result of reduced road congestion in our numbers. That reduced road congestion comes about as a result of HS2’s forecast of transfers from road to rail.
Q58 John Thurso: If road congestion goes, are you taking it straight from HS2?
Lewis Atter: Correct.
Q59 John Thurso: What about rail freight?
Lewis Atter: As I said, this is one of the areas that this analysis does not yet address.
Richard Threlfall: It is one of the areas where it suggests the numbers are understated today because we have not had a chance yet to look at the freight impact.
John Thurso: It is quite a big plus or minus, is it not?
Lewis Atter: It is a big plus.
Richard Threlfall: It is a plus. It must be a plus.
John Thurso: Yes, it must be. Thank you.
Q60 Jesse Norman: Can I just ask you, Mr Threlfall, how many people did you employ on this piece of research from KPMG?
Lewis Atter: There were more than six full-time, yes.
Richard Threlfall: We think about six, and in addition, as I think we have said in the report, we subcontracted Connected Economics, and there was an individual from that organisation as well involved in the work.
Q61 Jesse Norman: That was included within the fee you charged, the additional subcontracting?
Richard Threlfall: It was indeed, yes. That is included in that figure.
Q62 Jesse Norman: How many days do you think of work was done on this report by your team?
Richard Threlfall: The work extended over approximately four months from when we started, which would be in April, through to when we concluded and started to wrap up the work at the back end of August. It would be possible to go away and give you exact details from our timesheets on hours and so on, if you wish.
Q63 Jesse Norman: Did you have senior people involved in it, or were they all junior people?
Richard Threlfall: We most certainly had senior people. We had a lot of senior oversight in the work. It was important to the firm that we looked as carefully as possible at the results that we were producing. I had oversight and so did other colleagues in the firm.
Q64 Jesse Norman: Would you describe this as a small, medium or large-sized project for your organisation, in terms of this kind of analysis?
Richard Threlfall: In fee terms, it is relatively small. In terms of, I suppose, its impact, it seems to have had quite a lot of discussion.
Q65 Jesse Norman: In terms of the commitment of time for this kind of analysis, is this on the high end of the amount of time you would spend to do a piece of economic forecasting?
Richard Threlfall: On this sort. I think, as Lewis has already said, there is more work to be done, but it was a very intensive piece of work over the period of the four months that we did it.
Q66 Jesse Norman: I would be grateful if you could write to us with the details of how—
Richard Threlfall: We could certainly do that.
Q67 Jesse Norman: I am grateful. My question is, how can you charge as little as £200,000 for such an enormous piece of work over four months, involving lots of senior people, with junior support? It seems to me an extremely low number. Or were you pricing your partners’ time at very low levels?
Richard Threlfall: We had a long debate when we were competing for the work, because it was competitively tendered, back in March/April. We took a view as to the fee that we thought would secure the work for us. It was a piece of work that we were intellectually interested in, and I think it is one of the challenges of our organisation that we are intellectually interested in some of this stuff as well as trying to earn money from it.
Q68 Jesse Norman: You expected some future revenue from the client, potentially?
Richard Threlfall: Potentially. Equally, the team that I lead that does this, as Lewis well knows, is sold out across a number of pieces of work, so if we did not do this we would be deploying the same team on other things.
Q69 Jesse Norman: But you were really buying the work from the Government at a cheap price in order to do future work when they came onboard with HS2, were you not?
Richard Threlfall: No, I do not think we were buying the work. In practice, the rates at which we do this work will tend to be the same from one piece of work to another.
Q70 Jesse Norman: Good. We will look into that. Thank you.
Mr Atter, you caveated your answers to the Chairman by saying, “Subject to the questions we were asked.” What questions should you have been asked?
Lewis Atter: One of the challenges in all of this is that we are looking at HS2 in isolation. It is the impact of this project, using this methodology, in a future year. Some of the questions we might like to have been asked, had there been time and had it been a practical proposition, would be how this compared to alternatives.
Q71 Jesse Norman: All right, so you would have liked more option generation around what the alternative to HS2 would have been?
Lewis Atter: In an ideal world, yes, but the problem is that is quite a challenging question. This is a project that is delivered over 20 years. What you would be asking in terms of alternatives is what the alternative programmes would be over 20 years. Just to scale that for you, Government expenditure on capital runs at £50 billion a year, 3% of GDP. This is a 20-year project. While this project is being delivered on that 3% a year, total public expenditure on capital will be £1.2 trillion. To ask how this compares to the alternatives means asking how this compares to more than £1 trillion of as yet unspecified investment.
Jesse Norman: All right, I understand.
Lewis Atter: Major challenges.
Q72 Jesse Norman: Were you in a hurry to complete the work?
Lewis Atter: We were interested in moving as quickly as was prudent, in part because there needs to be a debate about this.
Q73 Jesse Norman: Do you think there has been adequate debate if parliamentarians get this report two days before a vote?
Lewis Atter: Publication date was not ours, so we did not know that vote was coming. The date for publication was not ours.
Q74 Jesse Norman: The answer you are giving is essentially no. It would be impossible to assess this piece of work, which took you four months to put together, in two days and reach a responsible decision on the merits?
Lewis Atter: It would be a challenge, yes.
Q75 Jesse Norman: Thank you. On the benefit-cost ratios that you have offered, am I right in thinking that they require these wider economic benefits in order to hit the targets you wanted? Is that right?
Lewis Atter: We are not doing benefit-cost ratios here. A fundamental point you need to recognise in all of this is this is very, very different from standard Department for Transport Green Book cost-benefit appraisals. All right? This is asking not about time savings. This is asking about the productivity and output in the economy, relaxing a lot of the restrictive assumptions that the Department for Transport makes. For example, the standard approach of the Department for Transport holds land use fixed. A transport scheme, whatever it is, will not change where people live, will not change where the jobs are, and will not change what kind of jobs they do. We relax all of that. We do not count the time savings. We focus in on what we think is the fundamental first order issue, which is impact on GDP.
Q76 Jesse Norman: Would this proposal, without that wider economic benefit analysis, pass a Green Book evaluation?
Lewis Atter: It is very different from the Green Book.
Richard Threlfall: It is different from the Green Book. If I may address what I think might have been a confusion—
Q77 Jesse Norman: No, stop. I am asking a question, a follow-up point to the previous question, of Mr Atter—as a former Treasury official, he can answer this question—whether this would have passed a Green Book appraisal as was without this extra thing. We can then come on to the additional point.
Lewis Atter: It is not a Green Book appraisal. It was never designed to be a Green Book appraisal.
Q78 Jesse Norman: All right, but the reason why you have done this is because of a concern that the Green Book approach would not capture the benefits as you see them, or as they were suspected to be, of the project?
Lewis Atter: That is fair.
Jesse Norman: Potentially the costs as well.
Lewis Atter: We have not looked at the cost side of the equation.
Jesse Norman: You have not looked at the costs?
Lewis Atter: No.
Q79 Jesse Norman: No. The effect of this approach is, is it not, to make this project non-comparable with any other project, because for no other project has this analysis been done?
Lewis Atter: I think it is true to say that no other national project has been looked at in this way. It is not true to say that projects on a regular basis are not looked at like this. If you ask how any of our major cities address the impact of transport and the benefits and costs of transport, whether it is a Greater Manchester, a Leeds, a Sheffield, a Birmingham or a Glasgow, all of them are increasingly looking at transport through this real economy lens. It is not the case that this sort of approach is not widely used. It is widely used, just not by central Government.
Q80 Jesse Norman: You said that the £15 billion of future national benefit, in your judgment, is an understatement, if anything, given other sources of potential revenue, as you see the future revenue.
Lewis Atter: That is our current view.
Jesse Norman: Yes. We have had an independent estimate that the net present value of that £15 billion is £256 billion. That is to say the project returns something like five times its capital cost. Can you see why, since that is a spectacularly high number, there might be some concern as to whether or not that is true?
Lewis Atter: Yes, it sounds like a discounted cash flow calculation. It would depend on the details. Remember this is an estimate for one year to produce a discounted view of that—
Q81 Jesse Norman: All right, but you are not suggesting that the benefits end in year 2?
Lewis Atter: Certainly not.
Q82 Jesse Norman: So the central case for future projections is going to be the year at which you made the calculation?
Lewis Atter: Absolutely.
Q83 Jesse Norman: All right, so it is not inappropriate then to discount that as a sensible case in a fan of potential terms?
Lewis Atter: We have not run those numbers, but I would not be surprised if they were of that order.
Q84 Jesse Norman: All right. If it was £256 billion or something close to it, you would agree that that would be a spectacularly high return for a project like this?
Lewis Atter: It is a good return.
Q85 Jesse Norman: Yes, out of any proportion to expectations.
Richard Threlfall: I think it is indicative of the transformational nature of putting an entire new railway line through the country.
Jesse Norman: Thank you very much indeed, Mr Chairman.
Richard Threlfall: Mr Chairman, if I may just clear up what I think might have been a confusion in one of the earlier questions, our report was published on 11 September, and I believe the report that Mr Norman was referring to previously was the economic assessment and strategic case published by HS2 last week.
Q86 Chair: Absolutely right, and that is helpful. Thank you for that clarification. Just to clarify what you are going to send us on the work you have done, it would be very helpful if we could see a reasonable breakdown of the timesheets.
Richard Threlfall: Of course.
Chair: I realise that is something that to some degree you will be reluctant to divulge, but there will be aspects that a client might reasonably want to see when deciding whether to pay a bill. In this case, the bill has gone to the public finances.
Richard Threlfall: There is no sensitivity on our side to providing that information.
Q87 Chair: This has been before your risk assessment committee, presumably.
Richard Threlfall: It went through a number of risk assessment hurdles. That is absolutely correct, yes.
Q88 Chair: What you were weighing, presumably, was a relatively low initial return against the hopefully good publicity by being associated with a project of this type. You do not normally do work of this scale for a couple of hundred thousand, do you?
Richard Threlfall: I do not think we did weigh it up in those terms. The rates that we did this at are our public sector rate, so in that sense we estimated the time that it would take us to do the work and that was the fee. We were in competition, so we tried to work out how we could do the work most efficiently, and I do not think we quite anticipated the degree of debate that the report would create.
Q89 Chair: I just want to go back to some replies from Mr Atter where he, I thought very sensibly, set out the difference between the Green Book approach and the approach you are taking; the former based on trying to estimate time savings and safety, and the approach you took based on land values and wage rates, and the effect, therefore, by that means on overall GDP. But you have told us, have you not, that these estimates of land values and wage rates, and I am quoting again, page 83, “Do not have a firm statistical foundation.”
Lewis Atter: We are saying that there is no perfect way of addressing the exam question that we were posed in doing this work. We have very transparently set out what we believe is a reasonable, probably conservative approach to getting at the key questions.
Q90 Mark Garnier: Can I just step back from your report a little bit and ask a few more philosophical questions, if I may, and your opinion on infrastructure in general? My first question is, why is this taking 20 years to build? Put another way, does it need to take 20 years to build it?
Richard Threlfall: Clearly it was not part of our remit to comment on the time it takes to build, and the time it takes to build is no doubt a combination of physically how long it takes to work out and how quickly you want to spend the money, but ultimately that is a matter for the Government and High Speed 2.
Lewis Atter: Also the time necessary to secure the powers.
Q91 Mark Garnier: Let me ask another question then. Why do you think the Government has to pay for this, and why is it not being put out to private tender for infrastructure companies that may find this an attractive opportunity to source yield, which, as we know, at the moment is something quite difficult to find?
Richard Threlfall: That is a good question and I am sure there will be a lot of work still to be done about the extent to which some form of private finance can come into this, but clearly the financing of the project is a different matter from ultimately paying for it. Ultimately paying for it, which is really what this debate is about, can only be met by the taxpayer or the user in some combination. Again, that was not part of what we were asked to look at.
Q92 Mark Garnier: I appreciate that, but you are both experts. Mr Threlfall, you lead on the corporate finance infrastructure business at KPMG, and Mr Atter, you are policy, economics, funding and financing of this stuff. This is a £50 billion project. Presumably, there must be plenty of people in the private sector who would see this as a great opportunity. Why is this not being considered? Why do you think it is not being considered?
Richard Threlfall: I do not think the detail of exactly how this will be financed has been completely worked through yet. There is a lot more work to be done on that. In terms of magnitude, a project of this scale would be quite a big undertaking for the private sector to undertake, purely on its own risk.
Certainly, in all of the experience of work that I have done, as soon as you get over a number of £2 billion or £3 billion or £4 billion, then you really start to stretch the willingness of the private sector to do that, certainly through the construction period. If you look at what was done on High Speed 1, that was essentially in the end taken forward by the public sector and then sold back into the private sector once it was an operational project. That would be a perfectly reasonable consideration, option, for this project.
Q93 Mark Garnier: With the length of time it is taking to build it, as I understand it, you start it and you just roll relentlessly forward until you hit Stewart’s part of the world in Scotland, which is why it will take 20 years. Surely, if it is so valuable an asset for the economy of this country, you could build it in 10-mile sections or 100-mile sections or start at each station and build in both directions. Would that just not be a better way of doing it?
Lewis Atter: I think those are legitimate questions. We have not looked at them, but they are legitimate questions. One of the things I think the Growth Taskforce is looking at, for example, is in terms of the detail of how you deliver this, how do you maximise the value? Those sorts of questions I would see as being a legitimate part of that kind of conversation.
Q94 Mark Garnier: My final question. In your opinion, if it is worth £15 billion a year in economic GDP impact in 2037, then it must be worth doing it quicker. If it is that valuable a project, surely somebody must come to the conclusion it is worth doing it quickly, rather than having a 20-year build period.
Richard Threlfall: My personal view on this, Mr Garnier, is that, based on the work that we have done, clearly in economic and financial terms High Speed 2 is a good thing and we should get on with it. The way I tend to look at it is we should imagine ourselves in 2050, we should imagine what it feels like to be in a country with another 15 million more people in it—
Mark Garnier: Quite a lot of us will be dead by then, of course.
Richard Threlfall: —and you should imagine what it will be like trying to get around this country if we had not made this sort of investment.
Lewis Atter: My view on this is I started thinking about High Speed rail as an ex-Treasury official, and clearly I started in the sceptical camp, given its scale and cost. However, in time, the more I have looked at this, the more I have asked myself the question: on a 20-year view, do you see a comprehensive solution to the north-south divide that does not have something like this as part of it? I am afraid I do not. In the context of a 20-year programme, back to that earlier number, the total in public sector investment over 20 years would be to the order of £1.2 trillion. The question about this is not, “Is this the very best thing to do?” It is, “Would you find space for it within £1.2 trillion if what you were after was a comprehensive solution to the north-south divide?” Against that background—deep breath—yes.
Mark Garnier: Thank you very much.
Q95 Chair: Thank you very much for coming to see us this morning. We may want to see you again at some future date, but in the meantime you are going to send us some written material in the light of some exchanges. Thank you very much indeed.
Lewis Atter: Thank you.
Chair: We are going to take a five-minute interval and begin again at 10.28 am.
Examination of Witnesses
Witnesses: Professor Henry Overman, Professor of Economic Geography, Department of Geography and Environment, London School of Economics, Dr Richard Wellings, Director of IEA Transport and Deputy Editorial Director, Institute for Economic Affairs, and Professor Dan Graham, Professor of Statistical Modelling, Department of Civil and Environmental Engineering, Director of the Statistical Modelling and Economics Research Group, and Research Director, Railway and Transport Strategy Centre within the Centre for Transport Studies, Imperial College London, gave evidence.
Q96 Chair: Thank you very much for coming to give evidence to us this morning. Do I take it all three of you have listened to what we have just been hearing and that you were in the room?
Dr Wellings: I was not in the room.
Chair: You were not in the room?
Dr Wellings: No, I was not in the room.
Q97 Chair: Okay. Well, then, let us start with Professor Graham and then maybe, Professor Overman, chip in if you want, and then I will ask another preliminary question to Dr Wellings. In response to the questions that I opened with and a number of people followed up on, we have been told that the £15 billion figure per annum over the 60-year period is robust, that it is a forecast, and we have been told, it seems to me in a somewhat contradictory way, that the key statistical work behind it is reasonable and reliable. There was some discussion about its reliability. What would you like to say about that?
Professor Graham: Okay. I would not agree with that. I do not think the statistical work is reliable and I have a few key concerns with it that relate really to how we would deal with this issue if we were doing an academic project or even, indeed, a student project, not necessarily a research project. There are some issues that you would straight off want to address. One of them is that the way the work was done in the report it kind of attributes to connectivity all of the things that affect productivity without incorporating them in the model. If you have a model that only has connectivity as the key thing that affects productivity, you are going to attribute lots of things to connectivity that actually just happen to be correlated with it but that are not down to the connectivity itself. The first issue is it does not deal with what is called confounding, so that is potentially a problem.
The second issue is that it does not deal with the problem of reverse causality. What I mean by that is that we tend to put transport maybe in places that are more productive so the relationship between productivity and transport is bidirectional. Places that are more productive tend to have better developed transport systems; think of London, for instance. It does not address that problem.
I think the third issue is that this business of what they called spreading the elasticity is not done according to any conventional statistical approach that I can think of. The problem they had was a problem that is called equivalence of outcomes where what they wanted to know is how we apportion the connectivity effect across different modes, but the statistics would not let them do it because all these things are so highly correlated. My approach to that would be: do not try to do it. It would not be to try to use some strange approach to apportion or spread that effect.
Q98 Chair: Would you be prepared to set down on a piece of paper in as near as possible layman’s language the point about the equivalence of outcomes with respect to the coefficients?
Professor Graham: Certainly.
Chair: Because it seems to me that that lies at the heart of whether this £15 billion figure per annum is rubbish in, rubbish out, or whether it is something on which, as we were told earlier this morning, we should rely. That is absolutely crucial. The whole project, it seems to me, stands or falls at the moment on this piece of work and this piece of work relies very heavily on that piece of statistical analysis. If I am not correct, tell me I am not correct.
Professor Graham: No, I think that is correct. It does rely heavily on that piece of statistical analysis, but what I am saying in addition to that problem are the problems with the numbers that they have generated that they then want to spread in the first instance, in my opinion.
Chair: I have understood that point and we have those on the record.
Professor Graham: Yes.
Q99 Chair: Even looking at it on their own terms, I think we need as a Committee to have an analysis of those coefficients. Professor Overman?
Professor Overman: Yes, I would reiterate what Dan has said. The issue about the treatment of the estimates that they get is an important one within the context of what they are doing, although I have to say I do not think that that is the first order concern about the substantive prediction that you would get out. They apply this statistical procedure, which really is not a statistical procedure. That is the problem. They apply this procedure, which is essentially made up, that provides them with an estimate. It is something that really should not be being done in a situation where we are trying to inform public debate using a statistical analysis.
Q100 Chair: That is a very serious allegation. You are saying they have just made it up?
Professor Overman: If you are trying to get at the impact of something like connectivity or productivity using a set of statistical techniques, you need to play by the rules. You have to implement procedures that will give you the best chance of getting an accurate assessment of what a change in productivity will do in terms of a change of wages. There are two technical problems with what they have done and then there is a more substantive point about the economics of it. You are pushing on the technical point so the two technical problems were as follows.
Number one, they had these four connectivity estimates that they could not separate out the individual effects on so they looked at them one at a time. Looking at them one at a time will always bias upwards the impact of any one individual effect because you will attribute to good rail connectivity the fact that places with good rail connectivity also have good car connectivity. This is a technical problem of using partial correlation coefficients or univariate regressions where you should be using multivariate. That was step one, which is a technical problem where they basically said, “We cannot separate these things out so let us just look at them one at a time.” Looking at them one at a time will bias up each of the individual effects.
Then they said, “Now we have these individual effects, how are we going to combine them?” The way they combined them is completely non-standard and further biases upwards the effects. If you look at their report—
Q101 Chair: Can you unpick the phrase “non-standard” for us, Professor? What do you mean by “non-standard”? Is it creative and useful or not? We need to know whether they have done something that can help us look at these sorts of projects in future.
Professor Overman: Taking partial correlation coefficients and adding them up in the way that they did is just non-informative, I am afraid. That part of it that gets the £15 billion number from the underlying partial correlation coefficients has no statistical basis. Just coming along and proposing a solution that has no statistical basis really does not get us very far because we have no idea how to interpret those estimates.
Q102 Chair: I am sorry to keep interrupting you, Professor Overman, but is it right for a layman to conclude from that we have to do something so at least they have taken this forward and that can help us understand where the value might lie, or is this rubbish in, rubbish out?
Professor Overman: I do not think that aspect of what they have done is at all helpful. They are basically doing something that is completely non-standard that gives them an estimate of £15 billion. Now, when they share-weight those partial correlation coefficients, which I have already told you are problematic anyhow, the number reduces to £8 billion. I would much rather have seen the £8 billion come out as their central estimate within the way that they were approaching the problem. That would still be a large overestimate. It would be a large overestimate partly because they estimated each of these effects individually and then because of more substantive problems about the interpretations of those estimates that Dan has alluded to but that I have not yet addressed in my own personal comments.
Q103 Chair: Okay. Dr Wellings, you have written quite a bit about this subject and I have read it. One of the main points you have made, as I understand it and correct me if I am wrong, is that this project has been captured by vested interest groups.
Dr Wellings: Yes.
Chair: In other words, a relatively small group of people in the country and those representing them are pushing for something that will be paid for at the expense of the rest of the country. Is that correct?
Dr Wellings: That is right.
Q104 Chair: What is the evidence that you have to support that view?
Dr Wellings: In my view, it is a classic example of concentrated benefits versus dispersed costs. It is a logic of collective action problem, in the public-choice jargon. The main losers are dispersed taxpayers who face enormous tax bills of potentially £3,000 per household. Because they are spread out across the whole country, their incentives to engage in collective action to campaign, lobby and so on—political action—is very low, whereas you have some very narrow concentrated interests with huge incentives to get involved in the lobbying process, collective action process. For example, train companies that stand to potentially win a £7.5 billion contract to manufacture the high-speed trains; the local authorities around some of the stations who will capture these massive regeneration grants from the taxpayer. Obviously, there is also a potential with senior officials as well to see their own careers and status to be enhanced by High Speed 2 and the opportunities it presents.
A huge amount of evidence and the only concentrated opposition really is the people along the route and so you see concentrated pros and concentrated antis battling this out but you do not see the taxpayers getting involved, even though they are the largest group in terms of the amount of money they are going to lose. You see the various pro high-speed rail campaigns being funded by Birmingham City Council, Manchester City Council, and by some of the major train manufacturers. You have pro HS2 conferences being organised with HS2 Limited, the train manufacturers, engineering companies and the city councils there who are going to get the massive regeneration subsidies. I was surprised how well this HS2 fits the theory. I think it is an absolutely classic example.
Chair: Okay, that is a helpful introduction to your way of thinking about the project.
Q105 Andrea Leadsom: Good morning. I would like to focus a bit on the benefit-cost ratio. First of all, Dr Wellings, there is talk of the DfT’s own rules that you should not include the wider economic benefit in calculating a benefit-cost ratio, but obviously it shows a significantly lower BCR if you do not include the wider economic impact. Have you done any work on that yourself and can you tell us whether excluding wider economic benefits is essential or just a matter of choice in this project?
Dr Wellings: Looking at the standard literature on this, the wider economic benefits are generally fairly low. I prefer not to include them because they are so questionable and nebulous and they also have to be set against wider economic losses. You have something called dead weight losses. It goes beyond the direct cost of the tax bill. It is this extra cost to society due to the economic activity that is suppressed by the tax bill. It varies enormously depending on how the tax is raised. It is relatively low for things like VAT, relatively high for things like fuel duty. It is potentially extremely high if something is funded with Government borrowing under certain circumstances where the Government already has high debts and so on. As a rule of thumb, the wider economic benefits are usually more than exceeded by these wider economic costs from the dead weight losses.
Q106 Andrea Leadsom: Has any work been done by the DfT or by you yourself in terms of assessing the wider economic costs?
Dr Wellings: Well, it is very difficult to do because you never have the knowledge to do it. You cannot really tell the marginal pound in tax that is going to be raised, or the borrowing, so you just have to make this assumption. That is part of what the Treasury’s job is, to factor that into the equation, and that is where they usually demand a much higher BCR than one before a project goes ahead. Assuming that the people behind the project are probably exaggerating the BCR, that is probably one of the reasons why you want a very high number before going ahead, even before you get into the whole opportunity costs argument.
Q107 Andrea Leadsom: Just to be very clear, it is your view that the wider economic benefits should be excluded from the BCR? That would be the right way to do this?
Dr Wellings: I think that is right. The WEBs are generally derivative of the other standard elements of the BCR anyway. In terms of deciding between projects, I do not think it makes an enormous difference to exclude them given that they are so nebulous and difficult to calculate. Going back to the KPMG report, personally I think it is highly questionable that you can make reliable forecasts 24 years in advance given political risk, disruptive technologies, and so on.
Q108 Chair: That does not really answer it, does it, Dr Wellings, because you have to do something? That would say because you cannot forecast something we had better not do anything where the returns do not come in for a quarter of a century. It is difficult but we have to acknowledge that they have a challenge on their hands. We are asking whether they have made the best effort.
Dr Wellings: But you are deciding between different projects anyway. It is an opportunity cost argument, so that is the answer.
Q109 Andrea Leadsom: Okay. Professor Overman, the fact that this BCR has gone down in this latest report, should we take that as sinister or just a swivelling out in the wash? Does it matter that it has gone down?
Professor Overman: I take a slightly different view to Dr Wellings. I think the BCR and the BCR with wider economic benefits are two important pieces of information that we should take into account when trying to make a decision about £50 billion worth of expenditure because otherwise we are guessing. Otherwise we have no mechanism for thinking about how to spend our scarce resources.
I would make the following distinction, though. In the DfT’s own guidance, at least the version of it that I accessed last week, which is called the WebTAG Guidance, they suggest that BCRs should be presented without the wider economic benefits. Now, my reading of that is that they should be presented without the wider economic benefits in for the purpose of the value for money comparisons that DfT make when they are calling something very poor, poor, medium, et cetera.
My view on that is that you should take the 1.8 BCR that comes out of that and using DfT’s own guidance say, “That is a medium-value project.” Now, I have to say that I went back and looked at what it takes to be a medium-value project in the distribution of all the BCRs we have out there for transport projects and it appears that you can get in if you are in the bottom 25%. So I am not sure that medium is the standard way in which we would use that word, but I think the 1.8 is informative. Then you add the wider economic benefits on to that and report them as a separate component because they are our best estimates at the moment of the wider economic benefits, the productivity gains and so on. I would make a distinction between those wider economic benefits that are reported in the economic case and the KPMG report, which gives you these huge gains if you net-present-value them. That is because I think those KPMG estimates are very overstated.
I do take the number seriously. To come back to whether we should take the 1.8 credibly, my detailed involvement with HS2 Limited and the underlying analytical work they were doing around the BCRs ended, I think, 18 months ago. At that point, I found that they had done a reasonable case in trying to assess those BCRs and those wider economic benefits that appear in the economic case reasonably carefully. It is harder for me to assess the variations that have happened since. The predominant issue seems to be we valued the business use down so, therefore, let us increase the number of business users that we assume are going to do this and the cost benefit, those two changes, balance out.
Q110 Andrea Leadsom: What you are saying, Professor Overman, makes it sound as if it is just a finger in the air. Is it at all reliable? Particularly, is it at all reliable on a stand-alone basis? Is it not the case that it is only really useful if you are comparing 10 projects and you have a BCR with wider economic benefits assessed on the same basis for each of those three or five or 10 projects? Can this on a stand-alone basis be at all useful?
Professor Overman: Let me answer the two parts of that question. I do not think it is just a finger in the air exercise. I do not think it should be the only thing that we are taking into account when we make decisions about investments because otherwise there would be no need for you guys and you would just get me ranking the BCRs and making a decision. I think it is unfair to suggest that it is not picking up something. These are best estimates about what we think the relevant benefits and the cost of this project would be and, of course, they are hugely uncertain. They are uncertain but I still think the central point estimates are informative.
How you use them is to make comparison to other projects. I understood the DfT guidance, when it classes things as poor, medium, et cetera, to be doing that on the basis of a standard BCR. If you want to make the comparison on the basis of the BCR with the wider economic benefits included—here I am talking about those of the economic case—you should go back to the Eddington report, which did try adding in those wider economic benefits. I have to say I printed it off this morning to see where it would stack up. Interurban corridors that Eddington looked at in 2006, the bottom 25% it looks to me had a BCR including wider economic benefits that would be about 2.5. That does give you a feeling that BCR with the wider economic benefits in would put high-speed rail in about the bottom 25% of interurban corridors that we had and it would be even lower if we looked at all of the types of schemes we had on the book because I am only telling you the interurban schemes there. You are right, you should not consider it in isolation, you should think about it in a portfolio of things, but I think you are wrong to suggest that we should just ignore the number, that there is no informational content in it.
Q111 Andrea Leadsom: Okay. Professor Graham, in terms of the argument for HS2 it has moved from one of speed to one of capacity and overcrowding. Is that a reasonable argument bearing in mind that the West Coast Main Line is not the most overcrowded line? Obviously, you are there to answer the question about the benefit-cost ratio, but if one was simply looking at capacity and overcrowding, is the central necessity to look at where the greatest problem with capacity and overcrowding is, and why we are not looking at a high-speed line to solve that problem first?
Chair: Just an easy question there.
Professor Graham: Yes. I think I will not answer that because I am not an expert in capacity and overcrowding on the network. What I would like to say, just to add to Henry’s point, is that benefit-cost analysis has been established over several decades now. There is a strong theoretical basis for it and over the decades we have built up empirical evidence that has been rigorously reviewed both within Government and by academics and through the peer review process. That counts also for wider economic benefits, which are, if you like, one of the more recent additions to benefit cost. The thinking was there is something missing; can we measure them? None of this is perfect, but we understand the assumptions behind it. We understand the theory and the empirical evidence gets reviewed and it is there. It is perfectly clear. It is all there. We can all look at it. We can debate it.
The problem with the KPMG approach is that that approach does not have that history. It does not have that academic support. It does not have the theoretical support. That is the difference. When we talk about wider economic benefits and cost-benefit analysis, we are talking about something that is very specific that we have really thought about and that we have analysed and put up to public scrutiny, and I was heavily involved in that work. That is not the case with the report that we are discussing today.
Q112 Andrea Leadsom: Right. So what would be your advice to the Government in order to get that accuracy? Is it possible to redo the report, to do a further report and do it on the proper statistical basis that you and Professor Overman have outlined?
Professor Graham: Undoubtedly, the work could be done better, yes. My advice to any Government would be if you want to think seriously about whether investments are worth making you have to look at things like benefit-cost because we have to be analytical about it. The £15 billion may be true, but certainly nothing in that report indicates to me that it is. There is not the scientific basis for it, and I think we could do a better job.
Q113 Chair: Just to be clear, what you are really saying is that KPMG were set the wrong exam question; that they should have been asked a different set of questions, which would have enabled them to look at the cost-benefit analysis in a completely different way? In fact, they have hardly looked at the cost side. They have only looked at the benefit side.
Professor Overman: Sorry, I am not sure who you are addressing that question to, but I am happy to answer.
Chair: I was addressing that to Professor Graham but chip in.
Professor Overman: The discussion there we are having is about how you present these things. You can take the numbers that they are doing and try to figure out what they look like, and on a net present value basis they look too high. The point we are trying to make is that as a part of the debate that happened around Eddington and other debates at the time, going back 20 years, we felt that the traditional standard benefit-cost appraisal process was missing something, that Dan Graham and others went out and carefully tried to assess that by coming up with these wider economic benefits that are now part of the NATA Refresh. We have a mechanism for presenting those. I think that the difficulty with the KPMG report is that we got all of that and then there were still people out there that felt that for some reason that was underestimating the benefits. They were then sent off to look at the wider, wider economic benefits and they have come back with an answer that, I am afraid, seems to me to be way too large. That is partly to do with the approach that they adopted and the robustness and so on of the report.
Q114 Chair: We are not going to try to do this now in front of the Committee, but if Professor Overman and Professor Graham were locked in a room together for an afternoon, would you be able to come out with an agreed methodology on how this should have been done?
Professor Graham: We would be able to improve it, that is for sure, because there are certain fundamental problems with it.
Chair: It is very risky saying anything near yes to that question because if you could try to do that we would be extremely grateful. I am not pinning you down. I just think it would be of some use to have an approach from experts on how we might get to a more thorough and reliable reply to the key question.
Professor Overman: The one thing I would say on that is that taking into account the evidence that is out there I have suggested that if you applied corrections based on what we know from the literature, you would certainly reduce the estimated magnitude by a factor of six to eight. That is coming off of experience in the academic literature where we have corrected for some of the problems in this report.
Chair: I have a few more colleagues who want to come in.
Q115 Mark Garnier: We have heard this morning quite a lot of talk from you guys about the fact that there seems to be slightly creative methodology used in terms of doing this. What do you think is driving this approach, Professor Graham?
Professor Graham: I would not want to guess. I think it is partly perhaps that the people who wrote the report were not aware of the latest techniques in this area. It does seem to me that the approach from the very beginning was under-ambitious in a way in terms of the data it was prepared to deal with. The work that Henry and his group at LSE have done and some of the work I have done is based on much more extensive data and models that try to incorporate and do things a bit more carefully we might say. I think it is under-ambitious. I do not know why they took that approach. I have not been involved at any stage in the work, I have to say.
Mark Garnier: Anybody else? Professor Overman?
Professor Overman: It would be hard for me to say. It is either something to do with the client or something to do with the material that was provided to the client. My feeling with this is that at some point HS2, DfT or the peer review panel should have pointed out that there were several parts of the report that were problematic and not in line with current guidance. This can happen when you are producing this kind of evidence on a tight timetable, but when the client has been informed by their expert peer review panel that there are genuine concerns about this, I do not understand, for example, why the £15 billion number was allowed to go out there. Even within the report itself the sensitivity test suggested that accounting for that one problem alone halved the estimate. None of us were involved with the peer review of it, but I would like to see what advice was provided to HS2 and DfT about the quality of the report and whether that was fed back to KPMG and why it was not responded to.
Q116 Chair: Advice from whom?
Professor Overman: Presumably, DfT and HS2 when they commission a £250,000 report take internal advice on the quality of that report from their own experts and they take expert advice, as we have been told, from a peer review panel. I think you would want to see what that peer review advice was. If that peer review advice said, “Oh, this is all fine, somehow this is acceptable,” then we want to question what happened to make that peer review advice fail. If the peer review advice was, “No, this is unacceptable. This methodology is non-standard. We really should not be out there,” then I think you want to understand whether it was KPMG that said, “No, it is. We are not going to reply to this” or whether it was HS2 or DfT that just said, “Okay, we are going to publish it anyhow.” Without that information it is hard for those of us outside of the process to understand where the sources of the problem have crept in.
Chair: We will ask for it.
Q117 Mark Garnier: Dr Wellings, what do you think? A funny exam question, a funny process—dump the Green Book? What do you think? Is there some Machiavellian plot here?
Dr Wellings: Well, I do not want to make any specific allegations but I think there is a general danger—
Q118 Mark Garnier: But you feel there is a question about specific allegations?
Dr Wellings: Yes, but we need to be a fly on the wall or for a whistleblower to come forward, which I hope will happen. It would be interesting to find out what has really been going on. I think there is a concern with high-profile projects of this nature that it does become highly politicised and that will start affecting some of the economic case behind it. In theory, research should be objective and so on, but in practice it is not and it will be influenced by the politics of the whole thing, particularly with these kinds of high-profile projects with a lot of political stakes, powerful vested interests dependent on HS2. The way you had a big increase in the budget and you also had a big decrease in the value of business travellers’ time and yet miraculously the BCR was barely changed, I do not think that kind of thing is credible. Either they got it completely wrong in the previous business cases or they are getting it completely wrong now, but there does not seem to be any accountability there. I think they have lost credibility. Whether it is political influence we do not really know, but there has to be a suspicion there.
Q119 Mark Garnier: Yes, okay, that is very helpful. Can I just ask the three of you the question I asked of the earlier panel, unfortunately you were not there. It is, first of all, the twofold question, if you like. Why does it have to take so long to build it? Why is it a 20‑year project when presumably if you are approaching it from a much more commercial point of view and if it really is so valuable to the economy would it not be easier to start it and do it in 100-mile stretches or 50-mile stretches rather than just rolling it out in one great long ribbon? That is the first question.
The second part of the question is why does it have to be financed by the Government? Why is this a taxpayer financing? I appreciate, obviously, ultimately it is paid for by the consumer, but if this is such a fantastic opportunity why is there not a case that it can be funded by commercial interests, who presumably at the moment are desperate to find yield with very low interest rates? Dr Wellings, do you want to start on that?
Dr Wellings: Right. Well, in terms of the timescale, it is so expensive it would put a strain on the general transport budget to try to do it more quickly. Even now at the peak I think it is going to take something like 40% of the Government transport investments in the peak year, which is enormous. There also would be a danger with doing it more quickly. It would be such a huge project it could inflate costs because it would place strains on the construction sector and the engineering sector, a bit like the Jubilee Line extension when they had to get it finished ready for the Millennium Dome. A lot of the firms could name their price, basically. Because it would create so many strains on the sector, there would be a danger that it would inflate costs that way.
Why does it have to be financed by the Government? Well, in commercial terms it is a loss-making project. Maybe the fares will cover the operating costs. They have not with High Speed 1. That still takes an operating subsidy and Eurostar had to be bailed out by the Government as well a few years ago. Most railways are loss making in commercial terms. Technically, they are not economically viable. It might be partly because they cannot capture some of the wider economic benefits, but I think if you relaxed planning rules you could probably capture more of those than a commercial operation. Nonetheless, I do not see why private investors would get involved unless it was a rigged market and Government guaranteed revenues, like the nuclear power deal, for example.
Q120 Mark Garnier: Professor Graham or Professor Overman, do you have anything to add to that?
Professor Overman: The only thing I would say on why it takes so long is that the Eddington report, the Barker Review of Land Use Planning, the recent LSE commission on economic growth point to the fact that we are very bad in this country at strategic planning for infrastructure investment, about pushing that through the planning process. That will explain a substantial part of the delays.
The other thing I would point out that I think is as interesting a question is why it is going to cost so much. The costs in this country do seem high relative to costs elsewhere and I think there is something to be done on why the cost-benefit infrastructure provision is so high per mile of the line.
Mark Garnier: Interesting. Thank you both.
Q121 John Thurso: Can I just start with one question that is not quite relevant to the discussion but is very relevant to me living in the far north of Scotland? The current HS2 proposal, the Y, stops in the middle of England and well south, obviously, of Scotland. To what extent have the costs or benefits taken account of the possibility of building on that to go to Edinburgh, Glasgow or wherever? Has any account been taken of that potential or is all the work that is being done on the assumption that it stops where it stops?
Dr Wellings: I will make an attempt to answer this. Obviously, in the business case they include the time savings for travellers from Scotland who are going to have a slightly faster journey as they link into the Y network and so on. This is a massive danger with high‑speed rail in that, to some extent, some of these vested interests—once, say, Leeds and Manchester get the high-speed line, then Glasgow and Edinburgh will want one and Cardiff will want one. Because it is all being funded by the taxpayer, it is a political process; it is not a commercial process. Probably there will be a spur to Liverpool in lieu of massive regeneration subsidies and so on. That is the problem. It is going to completely distort the whole economic geography of the UK and it is going to potentially produce massive burdens.
Q122 John Thurso: Can I just test that? For example, high-speed rail is highly competitive in time with air if it is sub-500 kilometres. Usually, the total journey time, if it is sub-500 kilometres, is quicker to go by high-speed rail. If you look at carbon emissions from aircraft, they are majored around takeoff and landing and, therefore, the shorter the flight the more per passenger kilometre you have of emissions. If, therefore, you had a high-speed rail link that took Glasgow and Edinburgh in, you could see a tremendous amount of aviation between London, Glasgow and Edinburgh stopping, as happened between Paris and Brussels. You could see a carbon benefit and you could also free up a lot of runway capacity in London for international flights. Yes, that is entirely political, I quite agree with you, but it is also quite important.
Dr Wellings: I am sceptical about the carbon benefits to start with because high-speed rail uses huge amounts of energy. As the speed increases you use more and more energy.
John Thurso: Nuclear-powered electricity.
Dr Wellings: Well, exactly, but in itself the high-speed rail does not create green electricity so it means to make the statistics look good they will do a deal and buy the green power, but all that does not displace the fossil fuel power to another electricity user.
John Thurso: Not if it is nuclear.
Dr Wellings: Well, the marginal unit is likely to be gas or coal powered because you are introducing more energy consumption by using high-speed rail but in itself that is not creating any extra green energy capacity.
Q123 John Thurso: I think the answer to my question is no, nobody has really thought that through?
Dr Wellings: Well, they have. The Conservatives originally introduced the idea of a high-speed line as an alternative to Heathrow expansion. It was bogus because it was never going to make much difference at all.
Q124 John Thurso: Do either of the others want to comment on that at all? No, thanks. Very good. Can I then go to the question of alternatives? Can I ask you all, first of all, do you think a rail network is needed or important or should we take the pure economic case that has been made and just let it decline peacefully and replace it with roads?
Professor Overman: I personally think the case for rail remains reasonably strong.
Q125 John Thurso: Is that a shared view?
Dr Wellings: I would rather see a level playing field in transport so I would like to remove a lot of the distortions we have at the moment, which would involve phasing out rail subsidies, possibly using some of that money—
Q126 John Thurso: Would you want to phase out the road subsidies as well?
Dr Wellings: Well, road is completely different because you have this £40 billion a year coming in in taxation from fuel duty and—
John Thurso: Yes, but it does not cover the costs.
Dr Wellings: It does cover the costs. It more than covers the costs. A lot of the external costs are inflicted by the planning system rather than road use per se.
Q127 John Thurso: You would like railway to just dissolve and roads to—
Dr Wellings: It would not dissolve. Reports published in the 1980s did a projection on what would happen if you removed the subsidies. You find that the 20% of the network that carries 80% of the passengers would survive. You would lose a lot of the rural areas but you would not lose the London commuter services and you would probably keep the main intercity trunk routes. You would have for most passengers—
Q128 John Thurso: London would be fine; Caithness, Sutherland and Easter Ross might as well join Norway?
Dr Wellings: No, it is irrelevant in those sorts of areas anyway, largely. Most people use cars. It is 90%-odd cars, isn’t it?
Q129 John Thurso: Well, we will move rapidly on. Capacity on the railways is space, time and speed. Either of those can affect capacity in that you have capacity at night which can be used. Whereas passengers go by day, freight is at night. You have the actual available rail space and you have the speed at which you travel. The faster you travel the more space you require and, therefore, the fewer trains can run. Are we all agreed on that? Does anybody disagree with that?
Professor Overman: May I just ask you to clarify the question? Are we talking about the alternatives to HS2 for dealing with the—
John Thurso: What I am trying to find out is if any of you understand capacity on the railway and the function of speed in capacity. That is the question.
Professor Graham: I am not an expert on capacity in the railways. I really came here to talk about the statistics underpinning the—
Dr Wellings: As a general rule in increasing speed, beyond a certain point there is an optimum speed in terms of capacity because of the braking distances and so on, but that can be changed by signalling and new technology can bring things together.
Q130 John Thurso: No, I think you are wrong. If you go and look at the reports from just about anybody who knows about transport, they will tell you if you take the current fast trains off you get more capacity. As we are near to full capacity on several of our lines, if we do not take the fast trains off we will run out of capacity. I am asking you which of the options, if you like; what are the alternatives? Let us assume we do not have HS2 for all the very good business case reasons, Dr Wellings, you put. That means that we run out of capacity within the next few years on most of our railways. The way we deal with that is by slowing them all down to ensure we can put more trains on or by rationing the capacity or by charging more. What do you think?
Dr Wellings: I am sorry, it is not actually true that. It is only quite small areas of the rail network that are facing overcrowding problems if you look at it geographically. It is mostly commuter routes into London. As was mentioned earlier, the routes into Euston are among the least overcrowded. You have huge amounts more people standing on the Tube than on routes into Euston, by an order of magnitude higher.
My solution would be, as I mentioned earlier, to firstly get a level playing field, remove the distortions by phasing out rail subsidies, removing price controls. Price controls are a huge problem. For example, the saver fare creates a situation on a Friday night where the first saver fare train is crammed like sardines and completely over-packed. More flexible pricing can make a massive difference. You have 12-car trains that are leaving Milton Keynes at 8 o’clock but then at 9 o’clock they are down to four cars, so if you can flatten the peak there is a massive amount of extra capacity available. Yes, phase out the subsidies, get rid of the distortions.
John Thurso: That is the time bit.
Professor Overman: Let us talk a little bit about the analysis that has been done for the alternatives to HS2 because I think that is the correct way to get into the problem. Sorry, I was a little bit confused by your line of questioning. Let me make a broad point first. Shifting the goalposts to suddenly have this about congestion and capacity is not helpful because it suddenly gets us into a world where all we care about is whether people are standing. The benefit-cost ratio and the wider economic benefits are trying to say what the value is of the journeys that people are making. We have people standing all over the transport network on rail, so just suddenly shifting it to the amount of people standing in peak hour coming out of Euston, which is not that congested a station relative to others, I do not find that shift to say this is all about capacity useful. What we want to do is look at the value of those journeys and do a proper analysis that figures out what the value is and compares the alternatives. Now let us look at how far we have got on that.
Q131 John Thurso: Sorry, I am not talking about capacity in terms of people standing. I am talking about capacity in terms of actual cars on the railway line, which may well be a freight car.
Professor Overman: Well, you see, it is the value of the journeys that are facilitated. I am perfectly happy to try to add freight into the analysis of the benefits that come from this, but just suddenly saying, “We have a crowded railway; therefore, we have to do something about it,” Eddington was very clear on this. We will have capacity constraints right the way across the transport network by 2036. What we do in those kinds of circumstances is to assess the alternatives.
Now, if you go and look at the alternatives that have been assessed so far, I would make two comments on that. One is that the benefit-cost ratios are higher for the alternative to the Y. They give us benefit-cost ratios of around about 3, including wider economic benefits, not 2.3. The reason why at the moment we have the Government not wanting to do that is we are told something about capacity, which gets us back to this nebulous thing of we are somehow trying to address the number of trains rather than the benefits of the journeys that are being facilitated. The other part of it is the point of how much disruption is going to occur. If you dig behind those figures—this is a 170-page report put out two days or the day before Parliament was set to vote on this—you find that this estimate of 14 years of disruption is basically got at by taking the total amount spent, looking at other projects that are spending money and seeing how much those other projects have caused disruption. It is based on a disruption per pound spent measure applied to the cost of doing the alternatives. I was struck by the fact that the alternatives are at an early stage of development, so I do not think we know what the disruption would be from the alternatives. That rather assumes that nothing is going to happen on those lines between now and 2037, which I also find hard to believe, and that we could not put these things in parallel.
If you are asking me what the alternatives are, pointing to capacity is a little bit of a red herring, but I would also say that the assessment of alternatives to date is not particularly satisfactory, which is rather disappointing for those of us who have been pushing for the alternatives to be carefully considered for about five years.
Chair: It is a very interesting piece of evidence and I know that others would probably like to come in, but we have Treasury questions this morning at 11.30, which does not leave us very much time. We have one more colleague wanting to come in and I am going to bring him in now.
Q132 Jesse Norman: Let us start with Professor Graham. Your expertise is in looking at these wider economic benefits, is it not?
Professor Graham: Yes.
Q133 Jesse Norman: Is there anyone in the country who knows more about this issue than you?
Professor Graham: At DfT I would imagine there is, yes.
Jesse Norman: Outside Government, you are probably the—you can say yes. The point is it is a huge area of your expertise.
Professor Graham: I know about it.
Q134 Jesse Norman: Okay, good. There are two areas of wider economic benefits that have been in question. One is built into the modelling that we have seen from HS2 and the second one comes from the KPMG analysis, is that right?
Professor Graham: Yes.
Q135 Jesse Norman: Okay, good. There are two levels and there is a question about whether these wider economic benefits can be properly added to the base case benefit-cost analysis that has been done. The assumption that has been made by HS2 is that the first level can and we have had a serious question raised as to whether the second level has any comparability with the KPMG work. Am I right about that?
Professor Graham: Yes, that is correct.
Q136 Jesse Norman: Good, that is helpful. Thank you. If the wider economic benefits were excluded from the first analysis, that is the first tier of wider economic benefits, it would fail the Green Book assessment, would it not? It would be classified as medium or possibly lower?
Professor Graham: What is the BCR?
Dr Wellings: It is 1.8.
Professor Graham: It is 1.8, yes, so as Henry said earlier it is medium.
Dr Wellings: Wasn’t that for the whole wider—
Jesse Norman: The point is it would be well below; it would be at the bottom end of the 25%, the bottom 25% that Henry was talking about.
Professor Graham: I think that is the point Henry made earlier, yes.
Q137 Jesse Norman: Okay, good, thanks. Professor Overman, why did you quit the HS2 panel?
Professor Overman: Well, I felt that the change in status of HS2 from providing independent advice to the Government to being a promoter of the scheme meant that I no longer wanted to be involved on a consultancy basis with them.
Q138 Jesse Norman: Thank you for that. You have said that the KPMG report overestimated the wider economic benefits in your view by six to eight times. That implies a £1.5 billion to £2.5 billion benefit. Is that right?
Professor Overman: Yes, that is right but I—
Jesse Norman: Not £15 billion but £1.5 billion to £2.5 billion?
Professor Overman: I think that is right, but I would not view those as being additional to the wider economic benefits that get the ratio to 2.3. I would see them as an alternative estimate of the thing that is shifting us from 1.8 to 2.3.
Q139 Jesse Norman: So there are two problems. One is that they are overstated by a factor of six to eight times, and the second is they are not additive. They are not additive; they are as an alternative way of calculating.
Professor Overman: I certainly would not add them on top of the BCR including wider economic benefits. I would rather see them as an alternative estimate of the wider economic benefits that should be added to the standard BCR.
Q140 Jesse Norman: I understand. Thank you. Would you share that view, Professor Graham?
Professor Graham: Yes.
Q141 Jesse Norman: You would share it, okay. That is interesting. Can I just ask about the absence of alternatives? Of course, we have heard from KPMG that they in a way would have liked to have been asked the question about wider options, alternative options, and option generation. Of course, there are two levels to that, Professor Graham. One is increasing capacity by other means and the second is other approaches outside using the rail network. Do you share that view?
Professor Graham: It seems reasonable to me.
Q142 Jesse Norman: You, too, Professor Overman? Or do you think it is surprising there are not more options that have been developed when you consider the amount of work to be done, for example, in the Eddington report on looking at alternatives to rail and alternatives within rail to increasing capacity by other means?
Professor Overman: My understanding, if you went back and looked at the earlier reports on option generation, was that they do say that they considered road and other non‑rail and decided that rail had to provide the solution to the problems that they were faced with. My concerns on that would be several. I have already alluded to the fact that I am surprised at the fact that these are at such an early stage of development, particularly with respect to the disruption that would be caused. I think that more regard should have been given to pricing. It seems very odd to me to rule out pricing as a mechanism for dealing with the capacity when dealing with the capacity—or however you want to phrase this—delivers a low BCR project. I would have liked to have seen more done on pricing.
I also have to say that I would have liked to have seen us dealing with alternative solutions that just dealt with the congestion question and not the speed dimension of this. When they were asked to generate alternative options, they were basically asked to generate options that met two Government objectives, one of which was to increase connectivity across the network, the other was to deal with congestion on the lines. I would have liked to have seen a separate set of options generated that dealt with the congestion on the West Coast, East Coast and Midland Main Lines to allow us to assess whether or not just dealing with the congestion would be something we would like to do rather than necessarily getting the speed benefits.
Q143 Jesse Norman: Professor Overman, in the Economist this week it says, “The notion that HS2 would transform the Midlands and the north is nonsense. High-speed railways in other countries have mostly benefited central cities by alleviating the need for regional offices.” What is your view on that?
Professor Overman: I think that its likely impact on the north-south divide has certainly been overstated. I am not sure we know. There are reasons to think it might narrow the north-south divide. There are reasons to think it might widen the north-south divide.
Q144 Jesse Norman: Final question—and I would be interested in any responses from any of you. KPMG said to us that this was a sufficiently small part of the overall transport budget that it ought to be seen in the context of the whole and that, therefore, in some sense too much attention should not be focused on the actual basis of the analysis itself. Is that your view? Perhaps, Professor Graham, you can kick off.
Professor Graham: No, I think the analysis is important because it is informing the debate and it is really important that we get these numbers right and that they are scientifically rigorous.
Jesse Norman: Important the numbers are right and scientifically rigorous you say?
Professor Graham: Important that the numbers are done as best as we can do them to the highest standard.
Dr Wellings: Well, I would just make the basic point that it is not a small part of the transport budget. It is a huge part and in some years it will exceed the investment in the strategic road network that carries around three or four times more traffic than the entire rail network in the UK. Yes, it is going to be a huge chunk, massive opportunity costs, and clearly if this investment was spent on schemes with much higher benefit-cost ratios it would deliver in order of magnitude higher benefits for the country.
Professor Overman: Yes, I think if you are going to spend £50 billion on a project the fact that you are going to spend £1 trillion on other projects does not preclude the fact that you should be careful about the £50 billion.
Chair: Thank you very much, all three of you, for coming in this morning. I am sorry this has been somewhat curtailed but that is because we have Treasury questions immediately afterwards. I think the Committee’s interest in this is undimmed by the two hearings we have had this morning and the likelihood is we will be asking for more. We are very grateful to you in advance for the co-operation I know you will provide. Thank you very much.
Oral evidence: The Economics of HS2, HC 788 19