Project Everest

Proposed PROJECT

PROPOSED PROJECT: (Malawi) Micro-finance digital platform for Australian lenders/Malawi borrowers

This is a rudimentary moonshot design for long run competition, which I have attempted to tailor to current solar-based micro-financing operations in Malawi.

An issue associated with micro-finance is the inherently high interest rates. There is high default risk and monitoring costs are proportionally high. Whats more, we discovered today that there are already several institutions providing services in this market, which may make it difficult to scale effectively as a for profit, non-local company. 

My idea follows the intuition of Kiva, but in a for-profit context. It could also be approximated to a CDO or a "micro-loan backed security". 

Under this design, Project Everest would act as an intermediate platform that would connect investors (in Australia or abroad) and locals (in Malawi) taking out the loans. 

Advantages to the investor are that instead of conventional investments in Stocks, Bonds, Deposits etc.. where an investor can expect for example a 6%p.a return, they will have access to markets where return rates can be much higher. There is also a potential attraction to this for the investor, as not only are the potential returns higher, but it may "feel good" to invest one's money in something which benefits developing societies.  

Advantages to PEV are that we can entirely hedge the default risk, as the investor/investors will assume this risk. While there is no risk, we can not expect to make a large margin, but a zero-risk premium can be taken off the top of the overall return as an "enablement fee". PEV will also then be able to charge lower interest rates, while still incentivising external investment, making the financing more affordable, thus making us a more competitive option. If the venture is successful, it may also lead to us having a larger access to cheap capital, which will make it more feasible to scale the project, and provide more loans. 


I think this can be directly applied and tested with our collaboration with Energy team and their solar product, which is going to be financed through an affordable payment plan which will benefit us both. 

Although it may not be a process that can be implemented effectively in this project month, I think it can be something that could be very useful come December. The reason for this is that we have new information, which suggests that adoption of the solar-powered lights we are selling around Blantyre is likely to be aggressive.

The case is that we will need to run tests with over the next six months, and we definitely will not have the scale to meet demand at all. It was strongly stated by the village elders in the communities we are targeting that they want to implement this or something like this on a payment plan, and if the test is successful then I think that to achieve scale we could use such a model.

From the point of view of a social enterprise, I think that micro-finance has potential to do harm though it may also do a lot of good. So more affordable finance would be more attractive. 


An example of how this could work:

If I have some money which I don't use saved in the bank, earning 3.5%pa. PEV offers expected returns of 15% p.a, through investment into developing communities. (This is an example, and would be calculated with effective default rates, which have been shown to be very low on the Kiva platform). 

If PEV has 30 people asking for loans of 200 USD each (again, illustrative example only) totalling at 6000 USD, I could put forward an amount 2000 USD to be invested in these loans. I assume the risk, and demand 15% return p.a. Imagine three others do the same, such that PEV has no capital outflow.

PEV has no cost of capital, and no exposure to default risk, so can charge much lower rates of interest, say 20% instead of 30-35%. This undercuts the competition and makes us more competitive and facilitates growth. We pass on 15% to the investor, and make a risk free margin of 5%. 

 Obviously, this is a very rudimentary design, and if it is feasible, it will obviously be more complicated, and further on down the track, but I feel that it something worth looking not and working towards.

It is unlikely that it will cause a second GFC ;)


Requirements to test this model:

We will need to check the results of the first test that we are planning on putting into action this coming week, to scope out what sort of interest rates are repaid on these product-based loans. If all goes to plan, the loans will be repaid at around 56%, which does leave us a margin to work with. 

We will need to source finance for potentially over 1000 solar products, which each currently cost around 28 AUD each. 

We will need to scope out if we can lower costs in the supply chain, and assess if the margin on the product is appropriate if we can achieve some economies of scale in regards to delivery and information collection etc.

Proposed MVP:

We will need to validate the conditions of repayment through the test we will conduct in the next 6 months. 

To gain access to a lot of capital, we would encourage investment from our networks in Australia. I propose that it be advertised within the PEV networks of current workers/trekkers and past workers/trekkers, as well as anyone who has any involvement and idea of what PEV is aiming to achieve. I think that it should be proposed that one person can invest in one product based loan, with an initial investment of 14 000 MK or around $30 AUD. We can then promise a return value of $35-40 AUD (This can be costed efficiently if this is a pathway we choose to follow). This amount will leave PEV a profit margin with no capital investment. I would suggest that this could become a feature of the PEV website, with an investment portal, or something similar. It would be good if the investor would be able to see the amount of invested funds they had, and at the end of the 6 months, then they would see their balance, and either choose to reinvest or not. 

I also suggest that we encourage reinvestment into the venture, thus generating more and more income income over time with zero cost, so that the project can regenerate naturally every 6 months, growing in scale each time. 

I suggest that we will also need to scope out different supply chains which may be custom designed to fit our needs. It would also be good to source a similar product under the PE name, and contract manufacturing based on our own design. I assume this would be produced in China or India, but designed specifically for PEV with the PEV logo on it. This is a lot of positive branding, which will strengthen our reputation in Malawi.

We will need to see how the licensing and importing will work if we are going to scale up to 1000 loans and products in December.

If there is not sufficient investment sourced from the PEV networks, we can also easily advertise on social media. Once again, we will structure the advertisement as a way to invest in developing countries in a socially appropriate and sustainable way, which is not charity. 


I think a deterrent around charity for a lot of people is that only 50% of funds donated are efficiently purposed for change. Whereas if we can clearly illustrate that every cent of funding is used efficiently and will also generate a return, I think we may find that people would be more willing to invest, and even to reinvest.  


Please comment if you have any ideas or criticism. 


edited on 27th July 2018, 12:07 by Wade Tink

Andrew Vild Jul 3, 2018

This is a good *concept*. What I would love to see is a proposed experiment in terms of how we can make this actually happen.

What that then means is as moderators, we can mark your post as "experiment proposed" or "experiment adopted" (more points!)

How I see this working in terms of a "MVP":
- design a basic landing page.
- create a simple Facebook ad.
- get your first "investment".
- "invest" it in a Malawian through Energy Product (given it is a more proven concept at this point.
- you now have proof of concept, and probably some very interested onlooking parties to make it happen in a biiiiigggg way.

Reply 1

Kurt Michl Jul 4, 2018

Thanks Andrew,
I did intend this to be just an idea, but I'll take on board what you have proposed and have a crack at it and keep you posted if there are further developments on this idea.

Users tagged:

Reply 0

Kurt Michl Jul 4, 2018

Thanks Andrew,
I did intend this to be just an idea, but I'll take on board what you have proposed and have a crack at it and keep you posted if there are further developments on this idea.

Users tagged:

Reply 0

View all replies (2)

Andrew Vild Jul 3, 2018

Status label added: Idea/Concept Not Complete

Reply 0

William Lee Jul 3, 2018

Hey mate, completely agree with this as a method of scaling up.
Especially interesting is the fact that frontier market returns are largely uncorrelated with developed market returns (e.g. S&P 500 correlation with African frontier markets = 0.134 over 2017). This along with ample opportunity to extract alpha means that there is a strong argument for strong diversified volatility-adjusted returns. So, there is definitely a strong argument for collateralising loans and selling them off to potential investors.

But as you say, this is a longer term goal. The key to getting there, in my opinion, is to be able to model credit risk appropriately so that we can prove low default rates and offer lower loan rates to provide greater social benefit.

While you say that there are local microfinance institutions (MFIs) operating in Malawi, interest rates charged are enormous. Attached is a study from 2013 on interest rates offered by Malawian MFIs. You'll see that the APR of most of these loans sit around 60-80%, and this is without fees. Interestingly, whether the loans were offered by a bank or NGO didn't affect the interest rate levels, suggesting that a profit motive isn't why loan rates are so high. Rather, as you've said, it is largely about default risk and operating costs.

If we can crack the problem of reducing operating costs by digitising much of the credit modelling and payment systems, we will go a long way to disrupting this space for the benefit of Malawians.

Reply 1

Kurt Michl Jul 4, 2018

Thanks for the insight Will, I've looked further into this today but I'd be keen for your opinion on the reason effective rates are so high with these loans.
We have been running on the assumption that the default rate on loans is at around 15% of all loans.
Would the fact that that interest rates are so high not be strongly associated with a strong lack of information?
What I mean to say is that because there is very little credit history due to the fact that the vast majority of the population isn't banked (67%), along with the assumption that debt is still not that frequent (32% of population in 2014) and even feared, make that there can't be a lot of credit history for the vast majority of people?

Wouldn't this be why the APRs are so high?

Users tagged:

Reply 1

William Lee Jul 5, 2018

For MFIs, I think there are four components driving the interest rate being charged:
1) The underlying risk of the project
2) The uncertainty premium being charged due to a lack of information
3) The operational costs involved (sales & marketing, credit scoring, payments & collection, legal & admin, etc.)
4) Inflation

We can look to reduce the first three components:
1) By offering loans only for 'good debt' (i.e. one that helps people save money or build assets in a long-run capacity - as such, being positive NPV) as opposed to 'bad debt' (often consumption loans). This helps to ensure that loans will improve the financial circumstance of consumers, and thereby reduce the underlying risk of the project.
2) As you've identified, this is about a lack of information/credit history. How we leverage alternative data to predict this will be essential in casting the net of financial inclusion further in these communities.
3) Digitising much of the operations will reduce costs - currently, the business model of MFIs are largely ineffecient, especially with regards to sales & marketing and payments & collections. They are mostly reliant on field officers that go out to villages, try to assess credit worthiness and are also responsible to collecting payments in person. Implementing digital credit scoring and payment methods will go a long way to reducing costs.
4) If inflation is 5%, making a 10% nominal return on a loan means that you've only made 5% in real terms. Therefore, we have to look a inflation-adjusted returns. If inflation was 12% and we were making a 10% nominal return, we would actually be losing money.

Please see the attached file for my summary of my views on this topic. Let me know if you have any questions!

Reply 1

Kurt Michl Jul 5, 2018

Thanks Will, That is some useful information.

Reply 0

Wade Tink Jul 6, 2018

The nature of commencing to 'start-up' in these economies for an MFI would mean that, just like we are facing, there is a need to concentrate the lending pool because of the operational costs involved. As all start-ups need to start with a very niche customer segment and expand out.
This is the opposite of what any financial institution needs to reduce capital risk. This is why I assume (to be validated) that the interest rate is so high. Thinking of the beta risk of this 'market' is exceptionally high as a drought for example, specific to a cluster of rural villages which you are lending to, could be catastrophic to your capital.
It is only through ensuring the product is digital that you can achieve the spread of loans which has a chance of reducing the risk.

Reply 0

Kurt Michl Jul 23, 2018

After some more exposure to the startup environment, I really can take a step back and appreciate what you're saying here Tink.
However I really do believe that there will be opportunities around the infrastructures that have developed naturally.
Although drought would be catastrophic for a rural customer segment, there is definitely a customer segment that would be suitable to a similar (even identical) value proposition within Blantyre for example. We have identified that Nancholi isn't the only optimal study focus area, and that markets in Blantyre or Limbe could very well be interesting. This would require more research.

Take a look at these posts Tink, as they may give some more context to what I am saying.

All the best :)

Reply 0

Andrew Vild Jul 4, 2018


I definitely agree from a project timeline perspective that this idea is a part of the longer-term strategy. That said, barrier to getting proof that this pathway is viable is low. I am certain that there would be individuals even within the PEV/Crowdicity community who would be prepared to put down $5-50AUD "investment" for a solar product to be microfinanced out - with complete transparency on risk/benefit.

While this would be a biased test, it would still be interesting to see if individuals back this idea enough to want to "invest" in a high risk, relatively high return model in early stages.

With respect to reducing risk and increasing visibility of decision-making information, let's make a post and define what is missing, what needs to be collected and can our newly created USSD survey quickly and effectively bridge that gap? What other opportunities exist to get this quickly so we can make agile decisions based off this.

We have 3 weeks to get results while in-country, let's get into action mode.

Reply 1

View all replies (6)

Amber Johnston Jul 4, 2018

Worth a read

Users tagged:

Reply 0

JayP Jul 6, 2018

This is an interesting concept and there is much that could be viable to this approach.

I think one thing that I would be interested to know is to whether or not you have a selection criteria into how you distribute your borrowers into group loans. Will this be by geographical locations or something more fundamentally financially/economically motivation. i.e. farmers or people who are have similar perceived credit risk?

Another issue that might be faced with this option is to with the fact that 67% of Malawi is unbanked with an even higher percentage within your probable customer base. Have you thought about alternative data sources like phone data? This is an emerging trend that if done right could definitely set you apart from your competition.

Reply 1

Wade Tink Jul 6, 2018

In response to "how you distribute your borrowers into group loans" initially it would be concentrated loans based on similar demographics for which you would need to compensate the providers of capital with the requisite risk premium. This of course would be high but it's one of the key restrictions of commencing a start-up that you must start with a concentrated customer segment
Should you be able to successfully digitise this process and achieve scale across different customer segments it would have a huge competitive advantage to be able to pool, hopefully, negatively correlated lending pools and create a better risk/return trade-off.

Users tagged:

Reply 0

Wade Tink Jul 27, 2018

I'm going to propose that this gets labelled as a "Proposed Project" and moved to the ideas box. It would be great to see follow-up comment with your perspectives on this now that you have been in country operating in the environment for 4 weeks.

My immediate concern is Australian regulation around providing a financial product as you are describing. Options may be to base the business overseas or seek to understand legislation around crowdfunding with the new changes under the Turnbull governments support for alternative sources.

Reply 1

Kurt Michl Aug 1, 2018

Thanks Tink,

I have had a few thoughts about this.
I think a good baseline products would be to replicate Kiva's operations, but with a specific use for the finance; the solar product. We will have set out some good data in December from the solar/payment plan experiment so it could be safe and valid to use this same product to test this platform out. We would be able to quote repayment and default rates in our sample of around 50 and provide appropriate interest rates to investors.

I think also, for the interest of simplicity legislatively, it would be interesting to offer this platform in a context where the investor would not experience any return. Although I understand the idea is not to be providing a charity service, I think that this "investment" could be offered with some form of credit or loyalty reward, rather than monetary compensation. More simply even, the "investment" could be accompanied by some photos and some "feel good crap" of the impact the solar product they have financed is having to the investor. Here we would be targeting people who would want to donate to charity, but don't. We can promise them their money back (or most of it, depending on the data we collect from the current payment plan) and a momento.

Alternatively, the interest collected by the investor on their investment could be withheld and accumulated to the stage where they could "redeem" it in the form of a reduction on a project fee, or simply more funding towards the project (exactly like Kiva, where interest is accumulated and reinvested continuously and the investor can only ever withdraw the principal investment amount).

It could be offered to current alumni of PEV that they sponsor an energy product, with the expectation being that they receive their investment back with zero interest. However, they will receive credit towards their next involvement with PEV.

Ideally however, it would be good to work as an investment platform without these restrictions, as it would open up to a larger pool of people.

Reply 1

Wade Tink Jul 27, 2018

Status label added: Proposed PROJECT

Status label removed: Idea/Concept Not Complete

Reply 0

Rose Martin Aug 2, 2018

I'm slightly late to the party on this one, but I think this is a sick idea!!!

It's a great way to raise funds while also reducing the risk for us as a financer. However, do you think this is the most viable next step for projects in December? My instinct for this would be that yes, it would be super cool to try to test the viability of whether people in Australia would be willing to put money into something like this; but is it worth it if we still can't prove whether we can model credit risk effectively and actually reduce interest rates using things like alternative data, education etc.? It may be more useful to put resources into testing our proposed model for credit risk, or looking into the similarities and differences between problems faced by multiple customer segments etc.
For me this comes down to where you want to allocate time and resources, and it seems that this is a cool end-goal to potentially work towards, but we are no use to anyone in terms of social benefit or longevity of the project if we cannot actually improve the way that finance is offered. i.e. if everyone just defaults on all the loans, then what have we really created?

I guess the way I see it could work in December is if we used the funds to simply continue to do experiments to further our business model in terms of using alternative data or education etc. but then we may have issues with repaying people in Australia if all loans default.

Happy to have a further discussion about this and whether you think this is the best move in December (am aware you posted this a while ago and your opinions might have shifted on the matter), so feel free to reach out!

Reply 2