APRA shuts the CLF, tells banks to buy assets

[This is an edited version of a post that was first published on livewire, on 10 September 2021]

In a move that’s a bit more aggressive than I had expected, APRA just announced that they will close the CLF (Committed Liquidity Facility).  

The CLF will be reduced in four equal steps (1 Jan’22: $104 billion; 30 Apr’22: $70 billion; 31 Aug’22: $35 billion, and $0 on 31 Dec’22). It has the same endpoint that I expected (zero in 2023), and while the starting point is a bit higher than I forecast (104bn v. 70bn), a clever tweak made up for the difference.

In practice the reduction is a bit more aggressive as from 1 Jan 2022 the first 100% of the LCR must be comprised of genuine High-Quality Liquid Assets (HQLA), meaning semis and govts.  They can use the remaining CLF for their buffers.

In a surprisingly strong letter, APRA says that they expect “ADIs to purchase the HQLA necessary to eliminate the need for the CLF.” 

For the system as a whole, this won’t be a problem upfront.  According to APRA’s quarterly ADI stats, the aggregate Australian Dollar LCR for LCR-ADIs, excluding the CLF, was about 108% in Q2’21 (it was 132% including the CLF).  However, the aggregate hides some important differences.  We know from the various Pillar 3 reports that smaller banks tend to rely on the CLF to a much greater extent for HQLA, so they are going to go looking for assets. 

As a general rule, banks like to get ahead of regulatory change, so I fully expect that they will seek to end their reliance on the CLF well before the closure at the end of 2022.  This means that they’ll also be looking for assets. 

If the aggregate system wants to get to 125% without the CLF, they’ll need to find a little over $80 billion of HQLA. If they want to get back to 132% on the AUD LCR, they’ll need to find $116 billion of HQLA (not all of the $139 billion of CLF showed up as liquid assets stacks as reported to APRA).  Of course, some of this will walk in the door as ES Balances, as a result of RBA QE.

Regulation (which I think should to be changed) means that ADIs won’t necessarily hold all this extra HQLA as bonds — though I think they will buy some bonds.  The lack of revenue associated with that ES Cash makes it a bit hard to stomach, particularly for the smaller ADIs (who tend to have lower ROE hurdles).  

The flows ought to be: more bank paper issuance (bank spreads wider, and wider AUD XCCY particularly in 5yrs); more bank buying of assets, particularly Semis on ASW (lower 6×3 and widening pressure on swap spreads).

My guess is that banks will prefer semis in the 7yr to 10yr section. 10yr Semi ASWs have probably seen the wides for the moment… 

Leave a Comment