Select Committee on Public Accounts Minutes of Evidence

Examination of Witnesses (Questions 60-79)



  60. Why should it not assume after 400 there would be further sales?
  (Mr Plenderleith) Because the first question asked would be, "Does the Government intend to sell more beyond this 200?" and if the Government said, "Yes, but we are not telling you how much", the price effect would have been much worse.

  61. Is it not never say never? Are you completely transparent in your intentions with a view never to change your mind?
  (Mr Plenderleith) If the Government had said we will review the position in a year's time that would have been taken as a signal that there could be more sales to come, that the process was open-ended, and that again would have had a bigger price impact.

  62. Presumably markets realised that changes could be made. You are saying if we had announced 200 tonnes instead of 400 tonnes the price would have gone even further down?
  (Mr Plenderleith) Yes.

  63. Very interesting. Presumably everybody who bought from the first or second auction has made a margin?
  (Mr O'Donnell) Again it depends when they bought, whether they held or whether they sold thereafter.

  64. They have all by the 15th.
  (Dr Mills) The gold price now is down to 265-267 dollars per ounce so if they bought at those auctions and held them until now their gain would have been from the second auction around 10 to 15 dollars an ounce.

  Mr Davies: Which is less than the rate of interest if you had converted it to currency. All right. Thank you very much.

Mr Williams

  65. I want to follow on what Geraint Davies has just asked, I am not sure there is much difference. In a way the criteria of whether this has been a success or not is whether, if we wanted to, we could now buy back the gold with the currency we have acquired and broadly get the same amount of gold back. You have indicated you would make a profit of about 100 million on a sale of 1.3 billion over the period, so that is about 7½ per cent, 5 per cent a year over the whole cycle. Did the gold itself earn us—and this is ignorance on my part—any interest while it was in the reserves?
  (Mr O'Donnell) I am not sure I agree with your premise in the sense that for us the big thing is lowering the risk of this portfolio. Overall we are not sure about the return and do not expect any significant change there but it will reduce the riskiness. On the question of how much one gets on gold, you can lease out gold and you get a rate of interest which varies but it is around half a per cent, and we are investing our US dollar assets, say, in things like Treasury Bills where we are earning whatever the US Treasury Bill rates are—about 6 per cent—so there is quite an interest rate differential there.

  66. There are two starting points from which you could start judging the success, are there not? There is the rate at which the decision was taken, when it was round about 283, and there is the rate, according to my translation of Table 15, at which the first auction took place, which was 261. If you look at it against the time when you made the decision to sell, actually it would appear, since you look at the period before that—and the table runs out only four months ahead of that—to be reasonably constant in the 280s over that stage. I do not know if it had been at that level in the period before the table before us starts, but if you judge it against the time when we made the decision, we would appear to have made an enormous loss.
  (Mr O'Donnell) Like I say, I really do not think you should get into this because what you will come up with, when you look back at it, is that gold peaked at about 800, something like, and you get numbers of billions if you look back at the cost of not having sold at that point. It is always easy with hindsight to go back and say, "Should one have done things differently if one had known perfectly about prices?"

  67. I can understand you saying that you do not think we should look at it like that, but this Committee looks at things like that. If, instead of selling gold, you had sold the steel industry, we would want to know whether you got good value for money. It does not matter to us what the product is, the fact it is gold is irrelevant actually. I am completely with you on the idea of diversification, I am not knocking that, and we are not here to question policy decisions. Our question has to be about the efficiency of it and the effectiveness of it. That is why I say there are a couple of starting points, and the starting points do matter, from which we can make our assessment, because by one you are breaking even, which is fine, but by the other, the time you launched this as a decision after which the market changed quite dramatically, it might appear to be a considerable failure. In the same way you say, "We do not know what the future is like", none of us know what the rest of this graph would look like after the date of the announcement had you not put all the gold on the market in that timescale and in those volumes.
  (Mr O'Donnell) But one thing we can be clear about is that the risk characteristics of the portfolio with a lower proportion of gold are much better than the risk characteristics with a much higher proportion. So that is an unambiguous gain.

  68. How are these risk characteristics assessed? Can you let us have a note on that? It might be in the Report. I must admit I usually do my homework very thoroughly on PAC Reports but I did not have a chance to do it this weekend.
  (Mr Plenderleith) There is a very good summary in the NAO Report on page 22 which describes the measure of riskiness which is considered.

  69. That is okay, as long as I know where to find it.
  (Mr Plenderleith) Particularly paragraphs 2.27 and 2.28.

  70. This period is artificial in two senses, is it not? It is artificial in the sense we were down-loading and it is artificial in the sense that the Europeans have said, "We are not going to", so you have had two contradictory market influences at work and still we cannot guess, at the end of five years, which is the end of the European announcement, how the market is going to settle. I must admit that the suggestion is that it will be a low price for gold, which would vindicate what you say.
  (Mr O'Donnell) I stress that I placed my case for the sale of gold on the portfolio diversification point. I am not into trying to guess what the gold prices might be in five years' time.

  71. No, but to some extent you had to do that to make the original decision as well. However, I do not want to go back over the same issues. Coming back to the point Geraint Davies raised about the 40:40:20 shares, this is very neat arithmetic but neat arithmetic when you are dealing with such complex issues suggests there is a rather rough and ready thought behind it. How does it happen that the ideal mix in disposing of our gold works out very conveniently to 40:40:20? Are these just policy figures with any calculation behind them? If so, what calculation?
  (Mr O'Donnell) Basically the decision was to sell the gold and put the proceeds into our foreign exchange portfolio without changing the currency composition of the foreign exchange portfolio. So the real issue then is what determines the 40:40:20, which is what your question comes down to. That is an analysis. Again I would not want to claim that 40:40:20 was in every circumstance going to be right but for us it seems appropriate to take a rather long-term view of that. We do not say, "Well, there are issues in the US economy, let's rearrange it and start selling dollars", we purposely stopped doing those kind of things. We take a long-term, strategic view. In terms of what we want our foreign currencies for, we look at proportions of our trade with different parts of the world, we look at the ease with which these currencies can be used for intervention purposes, issues like that, and in the light of those sort of considerations we come to the view that 40 per cent dollars, 40 per cent euros and 20 per cent yen is about right.

  72. There is a strikingly absent currency—why is there no Deutschmark element?
  (Mr O'Donnell) Because there are no Deutschmarks.

  Mr Williams: Because there aren't any! As I was asking the question, I was getting the answer myself! That is right, because they made the mistake of going into the euro!

  Mr Davies: Where are the roubles!

Mr Williams

  73. In terms of the balance of our portfolio now, are you satisfied with the total diversity of currencies you are holding?
  (Mr O'Donnell) Yes, at the moment, because most people would accept those are the three big currencies of the world—dollars, euros and yen. As I say, some way down the track, 10 or 15 years, one might want to consider if these proportions hold up, and this is something we look at fairly regularly, but it is a long-term view and it is something we would only move away from if we had a fairly clear view that our long-term analysis was changing.

  74. My final question is timescale. Why this particular timescale? Coming back to my point about two contradictory forces, the Europeans holding for five years and we off-loading, to some extent we might have derived more benefit from what they were doing because they announced it after us (and I do not know whether there were consultations) or we might have revised it subsequently once they made that announcement. Might that not have eased the price of gold because now you have got our offloading quickly the amount we have in 18 months while their bolstering effect on the rate is spread longer than the five years. Had we been more patient and taken the long-term view you keep referring to, might we not have got a better price if we had synchronised, once we knew of our decision, and if we had changed our situation and said we are now going to spread our sales over a longer period so that we do not have too dramatic an effect on the market?
  (Mr O'Donnell) I think you have to consider the dynamics of the agreement with the European central banks. What happened was we made our announcement, then a number of other countries, it was clear, because we are all looking at the same analysis here, were considering similar sorts of things. In a sense they were considering sales as well and then the issue was could we get them all together to make an announcement which would help the market because there were rumours in the market that everybody would be selling all of their gold and that was depressing the price. The Bank very skilfully managed to get together a number of European central banks (because for a number of countries it is the central banks that hold the gold) and they came up with this statement saying their sales would be according to this profile. We got the best of both worlds in that we were able to pre-announce our sales and then faced with that they negotiated an agreement around our numbers to fit in the rest which helped to support the price.

  Mr Williams: They probably thought, "Here come the British wanting to buy our euros that no-one else wants to buy", but thank you.


  75. Before I call Mr Rendel I have got two points. Mr Williams raised this question of the calculation of risk and Mr Plenderleith directed us to 2.28 and 2.29. 2.29, frankly, is a statement of motherhood of risk to reward ratios and 2.28 makes an assertion with no calculation behind it about the relevant risk being 25 per cent lower than the proposed portfolio. I think it would be quite useful to have a worked example to show how that is arrived at.
  (Mr Plenderleith) We can do that[3].

  Chairman: The other point to make to you, Mr O'Donnell, is that twice now you have made the point about your aiming to get a certain risk profile and we understand that. What the Committee is interested in is that you did that in the least cost or least loss way. That is why the focus is on transparency, the focus is on timing and the focus is on method. Can we take it as read that we know what you are trying to do. The worked calculation will help us to understand that and we can focus on the question of how we got there. Mr David Rendel?

Mr Rendel

  76. I have got rather a large number of questions but I hope they can be relatively short. I am going to start with two slightly naive questions but naive questions sometimes produce interesting answers. The first is why gold and not platinum, diamonds or any other precious substance?
  (Mr O'Donnell) Because we happen to hold a very large proportion of gold in our exchange reserves. We do not hold any of those other commodities.

  77. Did you consider whether it would be better to put some of the money into diamonds or rubies or something else rather than other currencies?
  (Mr O'Donnell) Yes. As I said, basically our view about foreign exchange reserves is that they need to be useful. If you think about what you might use them for one of the great uses of foreign exchanges reserves is for intervention purposes and currencies tend to be a lot better than commodities for that.

  78. Why are you keeping some in gold, which is presumably equally useless in intervention terms, rather than some of it being gold and some of it being in diamonds, for example?
  (Mr O'Donnell) Because it has some different characteristics and in some very unusual states of the world you might think that gold has particular advantages.
  (Mr Plenderleith) Unlike diamonds and other commodities it is a monetary asset and it can be used for making international payments and it fills a role in some of the international payments arrangements. So that is why it is held in monetary reserves in a way that other commodities like that could not be.

  79. The second naive question is if all the central banks appear to be selling gold, who is buying it?
  (Mr O'Donnell) It is being purchased for use by the industry for jewellery, etcetera, by end users. There is a certain amount of buying by people who believe that over the long term it will be a good bet.

3   Note: See Evidence, Appendix 1, page 18 (PAC 00-01/107). Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2001
Prepared 19 September 2001