Autobid and the SM

Thought I’d start a new thread for this, although there has been recent discussion of the subject on the “SM rates” thread. The current paucity of loans on the PM makes it easier to analyse Autobid activity and it is quite clear that Fleet-footed Chancers are artificially throttling AB’s access to the SM. My analysis of the current Loanbook tells me that there should have been £304k in repayments credited to investors’ accounts on 26th December and £398k on the 27th (this excludes “Late” and “Default” loans and excludes Whole Loans). I had no more loans go “Processing” on these days, so it’s reasonable to assume that virtually all this money would have been collected. Overnight on 25th/26th (when 26th repayments would have been credited), AB only bid on one loan (Somerset 4) as it was restricted by the 2 day/50% rule on other unfilled loans; it bid approximately £44k. During the day (26th), four loans passed the 48-hour point and AB immediately filled them up, investing approximately £39k.

Overnight 26th/27th, again only Somerset 4 was available to buy (it being the only loan left on the PM) and it invested approximately £40k to fill up the loan.

Now unfortunately I have to postulate as I don’t know what proportion of investor money has AB activated. Gut-feel based on observation of PM buying patterns is that this is likely to be in the region of 50%. If I’m correct, then between 26th and 27th, AB would have had some £350k available to it, of which about £123k went on the PM, leaving about £227k that it could spend on the SM.

So what did it spend on the SM overnight (26th/27th)? Well the difference between the total SM spend (per the Statistics page, which is updated in real-time) between 10:40 last night and 6:00 this morning was only £39k. Some of this would certainly have been manual bidders (some investors are active late evening), so I doubt AB actually spent more than about £25k. From the “Loan parts” page, I can estimate with reasonable degree of accuracy that the sum value of parts available at par or discount (AB doesn’t buy at a premium) was about £3.88m. So if AB spent £25k, that would be about 0.6% of what was available, which is exactly in line with my own actual sales. For overnight 25th/26th, I didn’t note the total SM spend on the Stats page, but my SM sales were again about 0.6%, so I suspect the spend on the SM overall was at a similar level.

All this suggests that AB is not spending anywhere near its available cash on the SM and so is building up a reserve to spend on PM loans not yet listed. This is contrary to the belief I have seen expounded on this forum that Fiddling Coffers uses SM purchases to ensure that Autobodger accounts are fully invested at all times.

GSV3MIaC has suggested there is a link (positive correlation) between the money AB spends on the SM and what it spends on the PM. Although this might empirically fit some recent observations, I don’t really buy it as it would be a very poor business rule - to ensure optimal management of investor accounts, Financially Competent should use SM purchases as a means of balancing out peaks and troughs in available cash. My guess is that they adopt a much simpler approach by simply telling AB on a daily basis how much to buy overnight on the SM. Hence on 22nd and 23rd they realised they had a surplus of funds following recent early repayments and so bumped the figure up, but while they’re not in the office to manage more closely (weekends and Xmas) they take a much more cautious approach and stick in a small figure.  The next two nights (with zilch available on the PM, and another £476k in repayments tonight and £314k on 29th) will be an interesting test.

Admittedly, as others have observed, some Autobidders will have set their rates high so as to avoid SM purchases. But I don’t believe this is constraining SM purchases at the moment, as on such a scenario it is difficult to explain why SM sales could go to three times the normal level on both 22nd and 23rd December.