Nationalbanken, the Danish central bank, acting as a ‘banker’ to the Ministry of Finance, might start to post collateral on its swaps in 2013. The swaps are plain-vanilla interest rate swaps and a few currency swaps.
The decision to start posting collateral on its swaps has been underway since last year, as reported by Risk. Denmark will join Hungary, Ireland, Portugal and Sweden, who — according to Risk — already post 2-way CSA (credit support annex).
Here’s the statement from Nationalbanken, from its ‘Danish Government Debt Management Policy Strategy 2013‘ publication:
The central government uses swaps for management of the central government’s interest-rate and exchange-rate exposure. In order to limit any loss on a counterparty’s default, the central government only transacts swaps with counterparties that have concluded a collateral agreement. The central government’s collateral agreements are unilateral. This means that the central government’s swap counterparties pledge collateral if the market value of the swap portfolio is positive for the central government, but the central government does not pledge collateral to the counterparty if the market value is positive for the counterparty. In 2013, the central government will negotiate new bilateral collateral agreements (2-way CSAs) covering new swaps. The switch to bilateral collateral agreements could lead to better swap prices for the central government since swaps with bilateral pledging of collateral decrease the counterparties’ liquidity requirement compared with unilateral pledging of collateral. The 2-way CSAs are planned to be implemented during 2013.
Nationalbanken is mirroring other central banks. Bank of England, for example, started posting collateral in 2012. Here’s from Lisa Pollack’s excellent post at the time:
With any such derivatives contract, it’s a zero sum game. When marking the transactions to market, if one party is up £1m (“in-the-money”), that means the other party is down £1m (“out-of-the-money”).
In the normal course of things, the out-of-the-money counterparty would post collateral with the in-the-money-counterparty. This keeps everyone happy because it guarantees performance under the contract. The exact rules around posting collateral are determined by an agreement between them called a “credit support annex” (CSA). The majority of CSAs are “two-way”, meaning that both parties have to post collateral as and when they are out-of-the-money.
But, sovereigns never really went for that. Instead, they have “one-way” CSAs. They expect their counterparties to post collateral with them, but they don’t expect to have to post collateral themselves. Banks were, more-or-less, willing to put up with this when counterparty risk was less of a concern and things were going a lot better for them generally. Before, say, the latest wave of regulation that takes an especially dim view of uncollateralised exposures.
Regulations aside though, there has always been something of a funding problem with trades like these (with sovereigns) since banks tend to hedge their trades.
In the above, we show that the Bank of England has entered into a swap with a dealer, e.g. an interest rate swap to hedge rates exposure. The dealer does another trade, or series of trades, with the dealer on the far left of the diagram to hedge the swap with the Bank of England.
Some time later, the dealer is in-the-money on the trade with the Bank of England, and out-of-the-money on the trade with the other dealer. This puts the dealer in a really uncomfortable position — collateral has to be posted with the other dealer, but the Bank of England doesn’t post any collateral.
So why should central banks, acting on behalf of the sovereigns, start posting collateral? It’s cheaper. Again from Lisa:
It seems they primarily did it to get better pricing on the derivatives contracts. It’s quite simple — the costs to the banks of putting the swaps together for sovereigns rose. It’s more expensive for banks to fund themselves, i.e. to get that collateral to post to their counterparties. It’s also more expensive to have uncollateralised exposure in terms of regulatory capital. The banks have been passing on these costs to their sovereign clients.
The Bank of England therefore concluded that it was cheaper to start posting collateral, as it should make the prices they are offered come down.
This makes a lot of sense for Denmark, too. Nationalbanken has a lot of assets in could post as collateral cheaply, and since they give up very little — the funding option — it’s cheaper to post the collateral. The banks, on the other hans, receives collateral that they can post to their counterparties (if they hedge), for capital requirements, and so on.
One could think that in a real crisis — where the government can’t pay — it might not matter much to to have Danish government debt, but that’s a whole other discussion which isn’t really that relevant. That’s not the issue. The point is that the swaps will be cheaper, and the opportunity cost to the central bank is very small.
A turning tide for two-way CSAs? – Risk
Danish and Latvian debt offices weigh two-way collateral – Risk
The Bank of England gets economical with its derivatives – FT Alphaville