Is there bank collusion?

Well, if there is, it isn’t obvious from the data.

With all the discussion of new scrutiny for the banks, one item that hung around (especially following the Commonwealth Bank’s recent hike in home mortgage rates) is that the banks are engaged in some collusive arrangement on mortgage rate movements. The basic story is that they ‘take turns’ in moving rates up or down in response to announced changes in the RBA’s cash rate. But somewhat curiously I have never seen an analysis of that story to see if it was true or not.

In response to that I (well, actually not I but a hapless research assistant) went back and examined all interest rate changes since 1996 and painstakingly (again, not I but hapless RA) identified which bank moved first in response and by how much. The raw data (and reference sources) are here. But the following diagram that I (yes I actually did this bit) constructed, shows the story. (If you click on the diagram you can download a larger image).

You can see that there is no apparent pattern of taking turns. To be sure, over the period, the four major banks shared the ‘first mover’ honours (although not equally) — CBA (16), NAB (11), ANZ (7), Westpac (11). Of course, being first to move down would constitute being a ‘good player’ so if there was an issue in collusion it would be for upward movements. The diagram shows this evaluation (with the icons for upward movements above the margin line). Over the period, for upward movements the turns were — CBA (8), NAB (8), ANZ (5), Westpac (3). This can hardly be called sharing the load.

Now what is true is that most of the time, mortgage rates only moved in response to changes in the RBA cash rate. There were substantial periods where that was unchanged (e.g., from 2003 to 2006) and so it is unclear why mortgage rates did not move for other reasons. Also, what is true is that it was (almost) never the case that a bank moved downwards with any lag but when rates rose, up until recently, the first mover took or two days to move. This indicates some reluctance to be the bad player.

It is also instructive to look at when rates were not passed fully through. This happened mostly for rate declines but there were some examples of RBA rate increases being followed by a greater pass through — last month being the most notable. Of course, that lack of pass through has been a post-2007 phenomenon — exactly when the non-major mortgage industry disappeared. Since then margins have been significantly higher and sadly, government interventions in the RMBS market have not stopped that trend.

Anyhow the data is available for others to look for a pattern or a conspiracy. For now, I can’t see it.

That does not mean, of course, that there are not ample reasons for renewed government attention on this sector — let’s face it, the analysis shows that competition is not working and I didn’t even look at business lending where things are potentially worse. It is just that this search for a headline conspiracy is a distraction.

4 thoughts on “Is there bank collusion?”

  1. Joshua
    Just a quick point in relation to this comment:
    “there were substantial periods where that was unchanged (e.g., from 2003 to 2006) and so it is unclear why mortgage rates did not move for other reasons.”
    The key reason why mortgage rates moved in tandem with cash rates — and didn’t move when cash rates were unchanged — pre-GFC reflected the fact that cash rates were the sole driver of changes in banks’ funding costs, which is mainly bank bill rates.
    These parallel shifts of the yield curve have, I think, led many to believe that mortgage rates are “determined” by cash rates, when in fact mortgage rates are influenced by banks’ funding costs.
    The disconnect that arose between BB rates and cash rates during the GFC-period has led to a disconnect between movements in mortgage rates and cash rates. This has come as a surprise to a public conditioned on believing that mortgage rates and cash rates must always move in lockstep.


  2. “Over the period, for upward movements the turns were — CBA (8), NAB (8), ANZ (5), Westpac (3). This can hardly be called sharing the load.”

    Coming in late, I know, but I’m not following. Are you referring to the fact that Westpac looks like something of a shirker. Other than that, it looks like fair shares

    More importantly, the big question is how reliably the other banks followed. I’d be impressed by an anti-collusion story if you could point to cases where one bank raised interest rates and others did not, particularly if the first bank then decided to back down.


  3. What about listing prices on foreclosed homes? There are whole neighborhoods of Los Angeles that are becoming bank-owned, and the listing prices are as if it is 2007. The house is obviously vacant, or staged professionally. The lawns are maintained. Also these homes are only getting one For Sale sign at a time, so as if to protect the neighborhood from the spectre of mass exodus.
    Are all the banks in collusion? Surely after 9 to 12 months some kind of correction to real estate prices should obviate? But they aren’t, and there are no multiple offers on these homes. Why aren’t the banks more willing to move these liabilities off of their books? Are they colluding, to try to force the government to bail them out again?


Comments are closed.