Speed of lending; cash drag

Following on from the previous discussion here are the details of my calculations of my average number of days to deploy.

Previously I said my average was 90 days and Steve posted his analysis of my data saying is was 32 days - quite a difference! I have identified 3 factors which I believe explain this gap. One was an incorrect assumption on my part, and 2 are to do with the way that Steve has calculated the average he has quoted.

Firstly a big difference is my assumption that deployment rate is dependent on the total capital on the platform. This came about because of a posting by Steve on April 12th as follows "The rate of deployment is constrained by the volume of loans not the value. So regardless if you invest £5k,£50k or £500k you would be at a similar percentage deployment - e.g. half way though." This led me to assume that if I doubled my capital on the platform I would get double the deployment rate. This seemed reasonable especially as my max loan size would also double. However if I now assume that deployment rate is dependent on the amount of uninvested cash on the platform, not total capital, this of course makes a big difference.

My calculation method is as follows. For each period, i.e. between dates where I added more capital, I hovered my cursor over the deployment graph on my dashboard and recorded my balance and the amount invested on the first and last day of each period. Then I divided the increase in investment by the number of days to get an average daily deployment rate. Using this rate I then calculated how many days it would be before the uninvested cash at the start of the period was fully invested. Simple enough! Can anyone see anything wrong with this method?

Doing this gave me the following deployment times : 48, 70, 25, 29, 31. Average 41 days. So why does my 41 days differ from Steve's 32?

Steve arrives at his time for full deployment by simply reading from my graph the date that the target was reached. This is not valid in my opinion because it includes a period following the addition of new capital that will have resulted in an increased allocation rate. Had I not added new capital then I would certainly not have achieved full deployment on the date he quotes. As far as I am concerned this method is completely invalid. It also has the curious property that money deployed between new capital being added and the old target being reached is counted in 2 separate periods, this can be seen from Steve's table above, Adding up his days column it comes to 160, but there are only a total of 117 days in the 5 periods in question. This seems like it must be wrong but I cannot explain exactly why.

A second factor is that I used a weighted average to calculate my average time to full deployment over the period I have been invested with BM while Steve quoted a straight average. Which is correct depends on which question is being asked. To illustrate take a simple example, a new lender adds a sum of money and this is deployed in 7 days, they then another sum and this is fully deployed after 48 days. What is the average?

If the question is "What is the average time to deploy a deposit " then (48+7)/2 = 27.5 is the correct answer.

If the question is "What is my average time to deploy during the 55 days that I have been lending on BM" then the answer is a weighted average taking into account the length of each period i.e. (7*7) + (48*48) / 55 = 42.7. This takes account of the fact that the bad period lasted much longer than the good period and is I suggest a better measure of how the lender feels BM has performed.

Using a weighted average on my first 5 periods I get a figure of 46 days.

Finally There are 6 periods in my investment profile, Steve ignored the last because using his method he cannot come up with a time to deploy as is has not got there yet. However using my method I can calculate it. OK this has been a particularly gruesome period over the summer holidays so perhaps not fair to quote it in isolation, but perfectly fair to include it in my overall BM experience I suggest.

So for my six periods my straight average time to deploy is now 62 days, my weighted average is 66 days. Of my six periods 3 are pretty well on the quoted average (25, 29, 31) and 3 are significantly over (48, 70, 166). I have had no periods "under platform average", so my overall average is well over the quoted platform average, I do not claim to be good at maths or statistics so perhaps my analysis is wrong or invalid in some way? I would very much like to have the opinion of anyone who is more competent than myself. Perhaps there is another, better method of calculating a reasonable average. A spreadsheet with my data is attached if anyone wants to check or modify my calculations. BM time to deploy.xlsx (14.08 KB)

Let me make it clear that time to full deployment/cash drag is not the most important factor with my perception of BM. Automatic allocation to quality loans with a low default rate is. I will be continuing with BM and will be adding more money as it comes off other p2p platforms with the expectation of getting good deployment rates as the summer holiday period ends. However we all know that the financial services industry is heavily populated by honey tonged sharks and charlatans whose only motivation is their bottom line and annual bonus, I have suffered at the hands of these people in the past, I intend to avoid it in the future. What I now need from any institution to whom I entrust my money, is total honesty and integrity, plus comprehensive and representative information about what I as a customer can expect. I completely reject Steve's calculation of 32 days as being in any way a valid calculation of my average time to deploy, for reasons explained above. I think it has no statistical merit and it is at complete variance with my intuitive feel for what BM has achieved with my money.

There are other complications in any calculation i.e. 95% or something like that could be considered as 'fully deployed' because repayments are coming of all the time, so this is not a simple thing. However the very best we can say about BM's performance on time to deploy a deposit is they that have done a very poor job indeed of managing customer expectations.

Edit : I do not want this post to seem too negative, so let me add that there are many aspects of BM that I like. It is just this one particular issue where they seem to be trying to defend the indefensible, Long term lenders who have indeed experienced the, mythical to me, 7 day deployment times are no doubt happy. However for us new comers things are just not working out in line with the expectations that have been set.

I appreciate the frustration that your experience doesn't correspond to the average experience. As discussed above, we are working on ensuring everyone's experience is closer to the average experience.

On your calculations:
 - you are double counting (carrying forward) the unallocated amount into each calculation - which is why it is over stated
 - your weighted average is irrelevant vs. the stated average performance

As we always say: the quality of investment is significantly more important than the rate of initial deployment (particularly if the hold period should be viewed as 12 months+, and we are talking about 1-2 months of initial deployment time).  Time and consistency will demonstrate this. We are now entering our third year of 8.0%+ pa. returns on invested capital and after defaults.  Which we are proud of and seek to maintain.

Loan activity is picking up after the summer lull.

Those who aren't fully invested should start to see their deployment rates ticking up.

I'm also pleased to report that after more than a fortnight languishing in the high 80s, my deployment has risen to 95% in the last couple of days!

I'd still like to think that existing funds being repaid took precedence in reinvestment over new/added money - what does anyone else think?

@yescautious ,

Wow, that is SLOW, even given that Steve F has admitted initial investments generally take longer than the quoted timescale, which apparently is a historic average of deployment time for ALL deposits!

SMALL is a relative term, as even the minimum investment is an average amount in my portfolio across several platforms - just to rub salt into your wound my deployment reached virtually 100% this week, (up 15% in the past fortnight, so I'm a happy bunny - congratulations to the team) so maybe it would be worth contacting BM direct to ascertain the reason.  Hopefully they will say they prioritise reinvestment of existing portfolios, and once you do get fully deployed you will benefit from that policy.  It would seem unfair to me if new money increased cash drag on established accounts, although feedback to date has suggested that is the case.

Pleasingly, I still have a low level of non-performing investments, which is my main concern - especially having just been informed of a mere 30% recovery of an allegedly secured loan on another platform!  Anyone else care to share their recent deployment experience?

Invested the minimum on 1 June 2017. Now at 70% so, rest assured, you're not alone.

As I am sure you know (and not saying that there aren't of course significant differences in the BM and AC offerings) Assetz have the state of the art system for this issue by sweeping into a lower paying account "while-u-wait". Makes a huge difference, both to a) overall return, but perhaps even more importantly b) how it feels to wait for full investment in the main account(s). Perhaps you could even do deal and use their QAA...

Anyhow, a larger issue for me than cash drag/deployment times would be whether an ISA is on the way - tax takes way more than a bit of deployment delay.