CBA warns house prices could fall 32%

Nice pickup from Dan Ziffer from the CBA update:

On those numbers CBA would only need $8bn in capital to replenish that which was lost.

I find that very hard to believe. It is probably derived from one of the usual garbage-in, garbage-out models.

David Llewellyn-Smith

Comments

  1. billygoatMEMBER

    YAY…the new first…as much as I’m living the house price / rent fall porn I won’t commemt YAY again. I will however enjoy it on the inside:)

  2. DominicMEMBER

    32% fall = $8bn capital impairment? Dreaming.

    Added to which, that qualifies as a ‘crash’ which would presumably precipitate outright panic – esp from investors who are heavily exposed and are ‘weak hands’. The correction might settle at 32% eventually but the correction low would likely be a lot lower than 32%.

    • Why do you think than $8bn won’t cover a 30% fall?

      Have you got back of envelope workings that you’d like to share?

      I haven’t done any me-self….

      Edit: I’ve done the numbers.

      CBA has a $700b mortgage book. It turns over every 7 years or so, so $200b has been written in the last 2 years and most at risk. (Remainder I assume to be fine, because equity mate, written at lower valuations, etc)

      Of the $200b from the last 2 years, say average LVR is 15%, so potential max loss is 15%x$200b = $30b.

      But $30b is lost only if 100% of the loans from the last 2 years default. If we say only one third of the loans default (one third is still a huge number), then the loss is just $10b.

      So $8b is not outside the realms of feasibility.

      • DominicMEMBER

        Wrong calculation: the value of the collateral is what underpins the security of the loans.

        This means the bank has to take a charge against all loans that are impacted by a decline in the value of the security. Obviously, that charge is based on the delta of the LVR – a bit like an option. The delta on a 32% fall in the value of ALL collateral will have the banks scrambling for capital. The ‘charge’ is meant to anticipate potential losses and if the losses are less than anticipated then banks tend to enjoy a good earnings bounce in subsequent quarters but if they’ve under-provisioned, oh dear … which is why they tend to over-provision in the early days of crisis (and bolster capital where possible). That or be nationalised / part nationalised).

          • DominicMEMBER

            It may sound sensible but it’s not the way it works in real life. Don’t get me wrong- the goalposts could well be moved by the regulator ie. a move to internal modelling or ‘Mark to make believe’. But investors are not stupid they’ll know (or guess) at the balance sheet damage.

    • Jumping jack flash

      Agree

      There’s a little over 2 trillion debt dollars as mortgages.

      The total value of the houses would be a bit more than that of course. Well, I’d hope so.

  3. Look, after Chris Joye has finished briefing the PM today he will give Matt Comyn and call and get this all straightened out. Might need to be tomorrow though, the Pope is calling for his daily update at 3:00 and he can just talk and talk!

  4. What a piece of garbage.
    So GDP in 2021 will bounce to 6% with 8% unemployment in a downturn.
    And, the worst case scenario in 2021, GDP will be -0.8% and 8.5% unemployment?
    So a -5.2% difference in GDP is 0.5% change in unemployment?

    • DominicMEMBER

      I discovered Excel just the other day – it’s amazing really, you can just input any numbers you like.

      Even better, you can create a ‘model’ and just change the inputs to generate the number you want! No wonder Bill Gates is doing so well.

Leave a reply

You must be logged in to post a comment. Log in now