Main index Financial Markets index About author

Switch Auctionettes

Julian D. A. Wiseman

Abstract: The UK Debt Management Office is planning switch auctions, in some form, to convert old illiquid non-benchmarks into larger more liquid benchmarks. These switch auctions would be more efficient (allowing a larger proportion of the source stock to be switched at more favourable terms) if the switch auctions were to be broken into a number of switch auctionettes, as follows.

Contents: Abstract, Publication history, Introduction, Recommendation

Publication history: This document is based on a letter sent (in the author’s professional capacity) to the UK Debt Management Office on 24th March 1999. Usual disclaimer and copyright terms apply.


On 9th March 1999 the UK authorities published the annual Remit, in which HM Treasury gives to the Debt Management Office its objectives for the year. These included:

¶24 The programme of conversion offers may be supplemented by switch offers into benchmark stocks during 1999-2000. Before the start of any such programme the DMO will publish proposals outlining the structure of such switch offers.

In the Debt Management Review, published on the 17th of March, the authorities expanded:

…the Government will be willing to consider additional measures, such as switch auctions.

and then expanded further in a questionnaire dated 18th March that was sent to GEMMs:

… [switch offers] are likely to involve a competitive bidding process whereby holders of a non benchmark stock bid a price (by ratio, similar to conversions) and amount to switch their holdings for a corresponding amount of a benchmark stock. It would not involve an attempt to reduce a source stock to rump status, but would essentially be a “top-slicing” process. It is doubtful [that the DMO] would consider amounts in excess of £3billion for any one transaction.


Switch auctions are an excellent method for the government to convert part or even most of an off-the-run illiquid stock into a strippable benchmark. There are risks, but these risks are avoidable.

In a typical government debt auction, one bids in cash for securities. Some governments allow non-dealers to bid (with different restrictions), but in practice the overwhelming majority of the paper on offer is sold to dealers. Because these dealers can borrow money against the purchased stock, dealers have access to effectively unlimited supplies of the ‘source stock’, which is cash.

This is not so for a switch auction. As an example, let us imagine the DMO has announced that it is to switch, by auction, £2bn of the old non-strippable £7.55bn 8¾% Aug 2017, into the strippable benchmark £5bn 6% Dec 2028. If dealers hold less than £2bn of the source stock just before the auction, and their bids reflect this, then the authorities will not be able to convert at reasonable terms. But if the dealers’ float is over £2bn, some dealers will be left holding source stock. Because of the auction, this source stock would be even less liquid — penalising those dealers who acquired it with a view to switching it. The squeeze could happen in either direction, and so a once-off switch auction would be risky for both the DMO and for the dealers.

Hence, a different regime is proposed:

Because of the small size of the switch auctionettes, it would not matter if they occurred on the same day as an economic release, but they should not occur at the same time.

The result would be a gradual and orderly winding-down of the issue size. As this winding-down occurs, the weight of the source stock in the bond indices will decrease, thus encouraging end investors to sell their (now overweight) holdings of the source stock to dealers, investing the proceeds into the destination stock.

Julian D. A. Wiseman, March and April 1999

Main index Top About author