Diversification: what is it really about?


One of the most spread finance principle among the overall population, and therefore p2p lending investors, is “diversification”. Many investors have this word constantly in mind while considering a new investment opportunity.

In the crowdlending world it translates usually into the following options: new platform, new country, new type of loans, new loan originator or new borrower from different economic sectors, etc.

However, do we understand the underlying mechanism of “diversification”? Do we actually diversify? Is diversifying “more” always a good idea?

Those are simple, yet important, questions for an investor. We all aim at maximising our benefits whilst minimizing the risk of losses.

This article will try to shed some light upon what really is “diversification” and, hopefully, make the reader doublethink whether adding more diversification in its portfolio is likely to be beneficial or detrimental to its performance.

What is “diversification”?

I am not going to invent a definition, you have plenty available online. Here is the one from Wikipedia:

“In finance, diversification is the process of allocating capital in a way that reduces the exposure to any one particular asset or risk. A common path towards diversification is to reduce risk or volatility by investing in a variety of assets. If asset prices do not change in perfect synchrony, a diversified portfolio will have less variance than the weighted average variance of its constituent assets, and often less volatility than the least volatile of its constituents.”

This definition highlights one aspect that is often overlooked by individual investors: “volatility” or “variance of the portfolio”. In other words, the risk is minimised by investing in assets that have a low price correlation with each other, enabling to minimise the variance of the portfolio and therefore the risk of losses (realised or unrealised losses).

Diversification is not what will make the investor maximising its absolute returns. The expected performance of the portfolio will always be lower than the most performing assets constituting the portfolio and likewise higher than the least performing ones. Diversification narrows the potential outcomes.

Somehow, the “cost of diversification” is accepting to dedicate part of your portfolio in assets with lower expected returns with the idea that, in case of underperformance of those having the higher expected ones, it will help counterbalancing your portfolio performance. However, this is true assuming that those assets have a low correlation among each other.

The application of the diversification principle into a portfolio management model has been theorised in particular by Harry Markowitz (Nobel prize in 1990), economist, who developed the Modern Portfolio Theory (MTP) in 1952. It enables to assemble a portfolio of assets with the aim to maximise the expected return for different level of risk. In such model, each asset risk/return relationship must be assessed depending of its contribution to the portfolio’s risk/return relationship, not on a standalone basis. The prices’ variance is used to assess the risk level.

How does it apply to crowdlending?

The objective of this article is not to get into details on portfolio management’s models. It is obvious that this cannot be perfectly replicated in the case of crowdlending. There is simply not the data to do it. In addition, the assets being unlisted loans (unlike listed stocks, bonds, etc.) the prices do not fluctuate depending on every single event happening on the market. The outcome are much more of binary type:

  • Will my counterparties (borrowers) honour their commitments and repay me or will they fail?
  • In case of failure, what is my recovery expectation?

For such reason, the term of “diversification” lose part of its actual meaning. It’s more about limiting concentration risk by lowering the amount at risk towards each borrower, platform, loan originator for a given portfolio size. A sense of correlation can be introduced in the selection of investments done by paying attention to the nature of the borrowers, their country of residence/activities, the riskiness of the underlying projects, etc. but it can’t be mathematically grounded.

When “diversification” loses its benefice in crowdlending:

I sometimes read that an investor has 20+ platforms and is still looking for the next one. I think the higher answer I came accross on social media was 40 platforms with about € 500 invested in each.

Besides the time consumption (p2p lending is often not really passive investment in my opinion) to manage so many platforms for a relatively low absolute return on each of them, is this a good idea at all?

Likewise, some invest in 40+ loan originators on Mintos. Notwithstanding that the actual diversification is lower than one can assume (ties between LOs and also with Mintos), is this really a good idea?

The short answer is: NO

But… why?

To base my reasoning I will borrow some terminologies coming from banks’ internal models. I will not get into details of those models obviously but I believe the analogy is interesting:

  • EAD => Exposure at default correspond to your portfolio
  • PD => Probability of Default (probability of getting a default)
  • LGD => Loss given default (in banking actually refer to the amount of default after recovery but in such case can also be assimilated to the amount of losses resulting of a default).
  • EL => Expected losses = EAD * PD *LGD (your actual losses)

I twist a bit the relations of the parameters but that’s the idea.

Assuming a given overall portfolio size (i.e. EAD), if you spread it among more platforms or LOs you will deacrease your exposure to each of them, therefore reduce the LGD, but at the same time it will increase the risk that one of them will default, meaning you increase the PD.

Therefore your Expected losses, being EAD * PD * LGD, might actually by higher if the decrease in LGD is more than offset by the increase in PD… this is when “diversification” becomes counterproductive.

It is rather straightforward that the more numbers of platforms or the more numbers of LOs you invest in, the higher probability there is that one belonging to your portfolio will default, ergo an increase of the PD. This is why conducting due diligence on the opportunity to invest in a given platform or a LO is MORE important than looking for spreading your money as much as possible.

Another aspect to take into account is that investors are humans being with busy lives and usually have a day job. Can the average investor dedicate so much time to properly assess whether investing in a given additional platform or LO is the right idea? The more numbers you have (i) the less time you spend determining that and (ii) the higher the probability you also include the most risky platforms/ LOs in.

The determination of those parameters based on historical portfolio performance and statistic laws is the underlying reasoning of internal models used by banks to set the level of provisions (as per international accounting standard IFRS9) to be accounted for in their P&L.

Those are also the core parameters (with some divergences in the way they are calculated) of models determining the risks beared by the banks (called Risk Weighted Assets = RWA) in front of which they have to hold capital. Provisions are there to absorb expected losses, capital is there to absorb unexpected losses (UL) which can come from events not captured by the models.

This explanation is meant to explain the analogy used above. But I am digressing a little bit, sorry. 😅

Conclusion: what are the takeaway?

Well, I obviously agree that putting all your eggs in the same basket is a bad idea but the extreme opposite is not much better as well. Therefore, here would be my (sometimes obvious) advices and things to keep in mind:

  • Do not think diversifying is the ultimate answer to risk management / safeness of your portfolio.
  • Diversify should not lead to overlooking intrinsic platform, LO or borrower’s qualities. These are important parameters to analyse to make well informed investment decisions;
  • Don’t diversify for the sake of diversifying, going too far can be counterproductive and leading to higher losses;
  • Keep your numbers of platforms and LOs manageable so you can keep an eye on the evolution of their quality;
  • By diversifying platforms and LOs you will have different counterparty risk levels. Ensure that each of them is aligned with your personal risk tolerance.

That is all from me on this topic. I hope it was useful to you and helped, if needed, to remind what diversification is really about. Applying it to p2p lending is everything but straightforward as there is no such data available as for stock markets. It has to be applied with common sense and tailored to each investor’s risk tolerance.

This is actually an additional take away, before investing: define your own risk tolerance and develop your investment policy/risk strategy based on it. 😉


Vues : 133

Catégories : Tips

1 commentaire

Leave a Comment