Model Portfolios, Equity Risk Premium and Total Cost of Ownership

Are off-the-shelf model portfolios getting more than their own fair share of fees?

The use of model portfolios have become common place within advice firms, with many choosing to outsource or use off-the-shelf model portfolios, typically provided by DFMs and research firms like Morningstar. And by off-the-shelf, I mean pre-constructed risk -rated portfolios, rather than bespoke discretionary management.

My bug bear is how these model portfolios providers tend to charge. Given that advisers fees/charging are coming under increasing pressure both from the regulator and clients, I think it’s time we put model portfolio costs under intense scrutiny as well.

My question is why are model portfolios charged on percentage basis? The fact is what model portfolio providers are really offering is their intellectual property, specifically risk mapping, asset allocation and fund selection. Why is that charged as a percentage of client’s asset under management? In most cases, the risk lie with the adviser to ensure that clients are put in the right portfolio, depending on their risk profile and so the risk to the model portfolio provider is minimal.

One key issue here is the Total Cost of Ownership to the client. Using outsourced model portfolio may add between 25bps to 50bps to client portfolio. Add to that the platform charge (30bps – 50bps), the underlying fund TER/OCF (75pbs – 100bps) and the adviser’s fees (75bps – 100bps), the total cost of ownership may be well in excess of 3%pa! Compare that to equity risk premium which historically has been around 4%pa (Well, 3.47% betwee 1977 to 2012, to be exact)

Does it not bother anyone that clients are paying 3%pa, so we could help them capture 4%pa real return? And what’s the implication of this for advisers’ long term relationship with clients? In these days where advisers are looking to build life-long relationship with clients, it may be a while but soon enough clients will figure out that they are paying so much, just to stand still.

The other way to look at this is, if for instance as firm you have £100M fund under advice and you hand 30bps of that to a model portfolio provider, that’s £300K pa of your clients’ money! Why not bring in the expertise in-house? I’m sure you could hire a good peer of hands for a lot less than that. Off course another alternative for firms is to run their own in-house model portfolio, perhaps with support from external party and we work with many firms to support the research process.

Finally, this is one area that adviser can use economy of scale. If you choose to outsource or use off shelf model portfolio, why not simply agree a fixed fee with the firm, where the provider design the solution, charges the adviser firm for it and then can pass the cost on to their clients? The more client you have in the model, the less the unit cost per clients!

I think as pressure mounts on advisers to demonstrate that they are giving value for money, the least we can do is expect the same of DFMs and other model portfolio providers.

 

Share:

12 thoughts on “Model Portfolios, Equity Risk Premium and Total Cost of Ownership

  • Interesting. We pay a fee for model portfolios so all our clients can use them with no charge at all, they just need to run them on the platform. It isn’t cheap to buy the IP, costs should be £20-30k per annum. It does throw up some other issues, however:

    1. As we are purchasing models to cover multiple advisers and we don’t mandate a platform, funds don’t always appear on every platform or in every product. Proxy funds sometimes need to be picked and whilst we help with that, advisers sometimes want something specific to their needs. Given it costs neither the adviser nor the client anything I think this is a compromise worth having. We don’t fetter the portfolio providers in anyway, unlike platforms who pay for them and often slant them to an in house view.
    2. As we don’t put an explicit charge on using the models I’m not sure all advisers treat them as seriously as ones they directly paid for. This sounds odd, but I’m convinced we would have more use of them if the costs were more visible.
    3. Many advisers take an use parts of the IP but not all of it. That’s how it should be, as we don’t make money from bps either. Some use the asset allocations, not the fund selections, others the opposite and some just use it as a start point for their investment committee discussions. For that reason it’s hard to track how many people use it and how much AUA is in it. I just know it’s well used and that justifies the spend.
    4. I’d suggest that model portfolio providers charging bps is popular in part as the cost is easily passed on the the client. It would be different if the adviser was paying directly. I do know a few firms, however, who reduce their fee to cover this charge or pay the fee to the portfolio provider out of their total charge to the client. At a certain scale I expect conversations on paying a flat fee will take place.

    Reply
  • Some interesting points made.

    However, of the ongoing charges you mention the only additional cost is that of the DFM, the others such as platform charges would be constant whether you use a DFM model or build a portfolio. So the question that needs to be answered is does the DFM add value for their 35-50bps?

    If they moved away from a % to a fixed fee no doubt the clients with smaller fund values would pay more than they are currently, and I think it would be fair to say it is the smaller clients that benefit most from these risk rated models.

    Finally, I do agree that the £300k could be spent on “in house” expertise, but the cost still needs to be paid for by someone (the client). The vast majority of advisory businesses don’t have anywhere near £100m in models; I suspect a tenth of that is closer to the truth and £30k doesn’t get you much as a “safe pair of hands”.

    Reply
    • Some good points Ian.

      First, the reason I singled out DFMs/Model Portfolios is that it’s one area I don’t see people shouting about. Cost of platforms, adviser and fund managers already under intense scrutiny by people a lot clever than I am. The lang cat platform pricing guide is a case in point, put platform cost under intense scrutiny. The folks Lipper are doing the same with fund managers. So, I don’t see the need for me to shout on these.

      I disagree with your points on small clients. They probably should be in model portfolios, if you consider the time it takes to rebalance and report to clients regularly, it’s not profitable for adviser. We encourage firms to use multi-managers for their sub £100K clients. But if they must, fixed DFM fee paid by the adviser can be passed on to clients based fund under advice, rather than per individual client. So cost is proportionate.

      Finally, average IFA we work with have about £20 – £30M FuM, but even at half of that, I can name more that a few ‘safe pair of hands’ to help with their investment propositions.

      Reply
  • Good article, but some comments

    Whilst costs do matter, they are not the whole story. We need to be careful in in measuring the value of advice simply against the charge against the portfolio, and resulting net growth. Advice should be more valuable than that.

    That said, personally, I feel 300+ bhps is far too high, and under 200 is more sensible, with many aiming for 150 or so. That takes pressure off the portfolio, and may allow for less risk to be taken as well

    I think both a BHP and a fixed fee can work, as long as the benefits of scale are passed onto the client. I agree on 100 million, £300 000 is high, but how many IFA’s really have £100 million on a bought in model portfolio? At that scale, you can do it yourself for far, far less. I suggest MP are for those with £20 to £30 million or so, and at that level, I suggest 15 to 20 BHP would be better. Please note, we do not use, or sell Model Portfolios at all, so this is based on what I’ve seen others using, and the threesixty experiences

    Lastly, Phils comments above are relevant. there is massive variation in the use of portfolios, from a cookie cutter / almost DFM approach, to a loose guide to Asset Allocation. So looking at a BHP for the latter seems wrong. A flat subscription to access data and IP makes sense

    Reply
    • Hi Phil,

      I agree wholeheartedly on value of advice – I am a bit of an evangelist on ‘Adviser Alpha’ – encouraging firms that we work with to focus on adding value to clients beyond investment returns, specifically tax planning, cash flow modelling etc.

      This explains why I think advisers (and clients) shouldn’t pay too much for investment proposition. Even at £20M FuA, that’s £60K at 30bps and I think that is still VERY EXPENSIVE for DFM’s IP.

      Reply
  • Hi Abraham, nice article. We’ve taken a different approach, for precisely the reasons you highlight above.

    We buy 3rd party research from Morningstar / Rayner Spencer Mills. We then run the portfolios on our own platform, at no charge to the client / adviser. The key difference is that we have our own discretionary permission, which allows us to aggregate trades (trade / switch costs are often overlooked in these costings for model portfolios and could add another 5bps in real terms to your numbers above) and largely automate the regular rebalance process, which we do on a “quarterly rebalance back to risk tolerance” basis

    It also means that the adviser does not have to contact the client for regular rebalances and fund substitutions. This has been a major headache for some firms I’ve spoken to who run their portfolios on an advisory / non-discretionary permission, as they end up running multiple versions of the same portfolio due to clients not responding to rebalance authorisation requests.

    We provide a full suite of research on the last day of each quarter, to be used for the following quarter – minutes from investment committee meeting, quant analysis, FinEx report for each portfolio, combined factsheets, commentary etc.

    We run 36 portfolios in total, all of them pretty vanilla but the total cost to client including platform charge is around 1% for active / 0.5% for passive. All our portfolios are risk rated and map to our in-house asset model.

    Advisers who are looking for something more sophisticated – direct equity and bonds rather than collectives for example – typically pay a premium of 25-30bps – although there’s downward cost pressure there too, we have one investment partner running replicas of their own model portfolios at 18bps inc VAT, which is more than compensated by the drop in TER for direct equity portfolios

    The vast majority of our advisers are Chartered / Certified and they have no interest in picking funds. They have no problem justifying their fees either! They generally run an objectives-based risk approach based on a detailed discussion / trusted relationship with the client and one client might typically be invested in 4-5 model portfolios, each representing a financial objective.

    We built our own platform exactly for this reason. It’s not available on the open market, but we now have around 2,000 clients in our in-house portfolios with an average account size of around £120k, and it takes us around 20 minutes per quarter to undertake the rebalances for all these clients.

    I absolutely agree with your sentiment – and I think that no client should expect to pay more than 2% for a model portfolio service, whether delivered in-house or via a third party.

    Cheers

    Keith

    Reply
  • Interesting. We pay a fee for model portfolios so all our clients can use them with no charge at all, they just need to run them on the platform. It isn’t cheap to buy the IP, costs should be £20-30k per annum. It does throw up some other issues, however:

    1. As we are purchasing models to cover multiple advisers and we don’t mandate a platform, funds don’t always appear on every platform or in every product. Proxy funds sometimes need to be picked and whilst we help with that, advisers sometimes want something specific to their needs. Given it costs neither the adviser nor the client anything I think this is a compromise worth having. We don’t fetter the portfolio providers in anyway, unlike platforms who pay for them and often slant them to an in house view.
    2. As we don’t put an explicit charge on using the models I’m not sure all advisers treat them as seriously as ones they directly paid for. This sounds odd, but I’m convinced we would have more use of them if the costs were more visible.
    3. Many advisers take an use parts of the IP but not all of it. That’s how it should be, as we don’t make money from bps either. Some use the asset allocations, not the fund selections, others the opposite and some just use it as a start point for their investment committee discussions. For that reason it’s hard to track how many people use it and how much AUA is in it. I just know it’s well used and that justifies the spend.
    4. I’d suggest that model portfolio providers charging bps is popular in part as the cost is easily passed on the the client. It would be different if the adviser was paying directly. I do know a few firms, however, who reduce their fee to cover this charge or pay the fee to the portfolio provider out of their total charge to the client. At a certain scale I expect conversations on paying a flat fee will take place.

    Reply
  • Good article, but some comments

    Whilst costs do matter, they are not the whole story. We need to be careful in in measuring the value of advice simply against the charge against the portfolio, and resulting net growth. Advice should be more valuable than that.

    That said, personally, I feel 300+ bhps is far too high, and under 200 is more sensible, with many aiming for 150 or so. That takes pressure off the portfolio, and may allow for less risk to be taken as well

    I think both a BHP and a fixed fee can work, as long as the benefits of scale are passed onto the client. I agree on 100 million, £300 000 is high, but how many IFA’s really have £100 million on a bought in model portfolio? At that scale, you can do it yourself for far, far less. I suggest MP are for those with £20 to £30 million or so, and at that level, I suggest 15 to 20 BHP would be better. Please note, we do not use, or sell Model Portfolios at all, so this is based on what I’ve seen others using, and the threesixty experiences

    Lastly, Phils comments above are relevant. there is massive variation in the use of portfolios, from a cookie cutter / almost DFM approach, to a loose guide to Asset Allocation. So looking at a BHP for the latter seems wrong. A flat subscription to access data and IP makes sense

    Reply
    • Hi Phil,

      I agree wholeheartedly on value of advice – I am a bit of an evangelist on ‘Adviser Alpha’ – encouraging firms that we work with to focus on adding value to clients beyond investment returns, specifically tax planning, cash flow modelling etc.

      This explains why I think advisers (and clients) shouldn’t pay too much for investment proposition. Even at £20M FuA, that’s £60K at 30bps and I think that is still VERY EXPENSIVE for DFM’s IP.

      Reply
  • Hi Abraham, nice article. We’ve taken a different approach, for precisely the reasons you highlight above.

    We buy 3rd party research from Morningstar / Rayner Spencer Mills. We then run the portfolios on our own platform, at no charge to the client / adviser. The key difference is that we have our own discretionary permission, which allows us to aggregate trades (trade / switch costs are often overlooked in these costings for model portfolios and could add another 5bps in real terms to your numbers above) and largely automate the regular rebalance process, which we do on a “quarterly rebalance back to risk tolerance” basis

    It also means that the adviser does not have to contact the client for regular rebalances and fund substitutions. This has been a major headache for some firms I’ve spoken to who run their portfolios on an advisory / non-discretionary permission, as they end up running multiple versions of the same portfolio due to clients not responding to rebalance authorisation requests.

    We provide a full suite of research on the last day of each quarter, to be used for the following quarter – minutes from investment committee meeting, quant analysis, FinEx report for each portfolio, combined factsheets, commentary etc.

    We run 36 portfolios in total, all of them pretty vanilla but the total cost to client including platform charge is around 1% for active / 0.5% for passive. All our portfolios are risk rated and map to our in-house asset model.

    Advisers who are looking for something more sophisticated – direct equity and bonds rather than collectives for example – typically pay a premium of 25-30bps – although there’s downward cost pressure there too, we have one investment partner running replicas of their own model portfolios at 18bps inc VAT, which is more than compensated by the drop in TER for direct equity portfolios

    The vast majority of our advisers are Chartered / Certified and they have no interest in picking funds. They have no problem justifying their fees either! They generally run an objectives-based risk approach based on a detailed discussion / trusted relationship with the client and one client might typically be invested in 4-5 model portfolios, each representing a financial objective.

    We built our own platform exactly for this reason. It’s not available on the open market, but we now have around 2,000 clients in our in-house portfolios with an average account size of around £120k, and it takes us around 20 minutes per quarter to undertake the rebalances for all these clients.

    I absolutely agree with your sentiment – and I think that no client should expect to pay more than 2% for a model portfolio service, whether delivered in-house or via a third party.

    Cheers

    Keith

    Reply
  • Some interesting points made.

    However, of the ongoing charges you mention the only additional cost is that of the DFM, the others such as platform charges would be constant whether you use a DFM model or build a portfolio. So the question that needs to be answered is does the DFM add value for their 35-50bps?

    If they moved away from a % to a fixed fee no doubt the clients with smaller fund values would pay more than they are currently, and I think it would be fair to say it is the smaller clients that benefit most from these risk rated models.

    Finally, I do agree that the £300k could be spent on “in house” expertise, but the cost still needs to be paid for by someone (the client). The vast majority of advisory businesses don’t have anywhere near £100m in models; I suspect a tenth of that is closer to the truth and £30k doesn’t get you much as a “safe pair of hands”.

    Reply
    • Some good points Ian.

      First, the reason I singled out DFMs/Model Portfolios is that it’s one area I don’t see people shouting about. Cost of platforms, adviser and fund managers already under intense scrutiny by people a lot clever than I am. The lang cat platform pricing guide is a case in point, put platform cost under intense scrutiny. The folks Lipper are doing the same with fund managers. So, I don’t see the need for me to shout on these.

      I disagree with your points on small clients. They probably should be in model portfolios, if you consider the time it takes to rebalance and report to clients regularly, it’s not profitable for adviser. We encourage firms to use multi-managers for their sub £100K clients. But if they must, fixed DFM fee paid by the adviser can be passed on to clients based fund under advice, rather than per individual client. So cost is proportionate.

      Finally, average IFA we work with have about £20 – £30M FuM, but even at half of that, I can name more that a few ‘safe pair of hands’ to help with their investment propositions.

      Reply

Leave a Reply