Guest blog: Paving the road to resilience

4 February 2021

Ask any lender and they will say that small-scale loans don’t come cheap – by definition. They are more cost intensive to administer and more likely to be sought by ventures typically seen as higher risk because, for example, they are early stage or have no collateral. Little wonder then even in the social investment space, few investors were responding to demand among VCSEs for affordable and risk-tolerant sub-£150k loans.

Market testing in the first half of the 2010’s had indicted this systemic block in the market sorely needed to be overcome - to enable VCSEs to access working capital to underpin their business models. That’s why The National Lottery Community Fund clubbed together in 2016 with Big Society Capital (BSC) to set up the Growth Fund – managed in a wholesale capacity by Access (The Foundation for Social Investment) - which would create a group of social lenders ready to make offers and support VCSE resilience.

Samantha Magne - Knowledge & Learning Manager at The National Lottery Community Fund – considers the early lessons from the Growth Fund evaluation produced by Ecorys/ATQ, about what these lenders make of how The National Lottery Community Fund’s subsidy, combined with BSC capital, has worked for them and the VCSEs they support.

So if running a blended-finance scheme is not quite your usual cup of tea – but you have a thirst for learning about how to get capital to the parts of the sector that need it most it, read on.

This is a story in three parts; the story of Grants A, B and C, all performing a trinity of jobs to get the pipeline of money flowing towards VCSEs who need finance to entrepreneurially generate new income or manage cash flow.

• Grant A is there to create a ready and willing group of social lenders, covering their operating costs over an initial ‘runway’ until they can get fully off the ground and sustain them themselves through returns and fees they make on deals.

• Grant B is there to redeem them from losing money when risker ventures default – as it can be used instead to repay their wholesale capital investors (up to a certain amount).

• Grant C can be passed on in good spirit by the lenders alongside the loan deals to their VCSE investees, which they do in multiple ways.

You’ll see why some kind of aide-memoire is needed to navigate the logic, virtues and drawbacks of this three-in-one subsidy arrangement and how it has created ups and downs for the lenders. Stay with me and we will take each part in turn:

Grant A (av £120k) was designed to give lenders a kick-start proportion of their operating costs - with the rest of their costs coming from a slice of the return on their investments. Grant A was designed as an acknowledgement that there’s no way the lenders could hope to cover all their costs simply by bringing in profit from a small slice of returns on investment – even with the potential to boost income generated by recycling any repaid principal capital into new deals.

As a starting position, the programme partnership provisionally allocated Grant A to cover an initial runway, at approx. 10% of overall projected op costs. There was some flex in this, recognising that newer lenders cautiously setting up smaller funds would need relatively higher proportions of Grant A than social investors who were already established in other markets.

Despite this best-guess modelling however, loan pipelines were slower than anticipated to convert into live deals and some defaults came earlier than expected. This made for a tight situation for most of the Funds and not least for the newer lenders (although less so for one who opted to support more mature organisations with larger- i.e. less risky - deals). The parent organisations behind the social lending schemes did expect to help smooth an element of cashflow shortfall and even to underwrite it, but some lenders found themselves having to call down extra capital from BSC to cover their op cost shortfall, and three had to cut their investment management team capacity down.

As a consequence, several were left having to downsize their overall loan deployment plans (- albeit a couple of established lenders were in a strong enough position to increase their fund size). This retrenchment only made their chances of making the margins to cover operating costs even tighter. The programme’s viability tests, including an asset coverage ratio, had been intended to act as an early-warning system, but the evaluators issued a warning - to the warning and response system!

In effect, once deployment falls behind plan, the viability tests built into the model may be generating their own downward pressure. [The model] does not allow for adding resources in order to catch up. Re-profiling discussions between social investors and the funding partnership have, for example, extended deployment windows but, either not allowed for additional costs or, demanded further operational cost reductions.

Around half the lenders were not due to wind down their active deal-making until late 2021 early 2022 - so in theory some could still make up ground. However this was looking unlikely even before Covid-19 arrived on the scene - and so its advent served only to reinforce the option (that was already under consideration) of re-distributing of the remaining subsidy funds between those lenders in a stronger position (to make new deals, keep recycling the capital and see existing investments through). 

The wider Covid-19 government response of bounce back loans etc. had also inevitably changed the demand in the market. It’s important to note though that GF lenders were widely praised by their investees for the expert support and flexibility they have offered VCSEs to help them cope. And, that 45 new deals were still done in the first 8 months of Covid-19.

Grant B was designed to enhance flexibility around risk - by helping lenders cover losses from defaults. When combined with the initially raised capital, it accounted, on average, for 29% of the funds the lenders had at their disposal. However Grant B was only issued to lenders at the time of - and in proportion to - their draw-down on their capital facility.

This arrangement served to keep the grant closely linked to active loan-making (rather than handed out as an unconditional subsidy) but, it meant that even though BSC don’t call in the capital repayment until the end of the scheme, some lenders began to worry that early defaults could leave them short in the meantime. Any early defaults should be mitigated by subsequent drawdowns in the fixed BSC/Grant B ratio. But the modelling is complex business. And some lenders had also hoped to retain rather than defray their grant B if all went well. 

So, although Grant B is intended to help lenders make riskier choices all the way through their funds, the evaluation has indicated that disbursing it this way can still dampen that effect, creating tendencies to push risk-appetite to later down the line in a Fund’s life, when the lenders’ ultimate liabilities can be more easily reckoned. (Most Funds had originally planned to stop making new deals before end of 2022).

It’s also worth noting why and how in this structure, and due to the rules that governs BSC’s remit, Grant B has worked primarily to help BSC preserve its capital and costs of curating its pot; BSC draws from dormant accounts capital and is tasked with making sure this is not eroded whilst also having to generate revenue from the capital to cover BSC’s costs in deploying and recouping it. But in the Growth Fund, BSC (and a few other capital suppliers) have no recourse to the lenders themselves. So BSC (and TNLCF as the Grant B provider) would ordinarily need lenders to pursue any defaults from VCSEs as far as possible. 

However Grant B helps to take the heat off this imperative to chase defaulting VCSEs. And that’s worth noting as we look back on our original intentions for supporting VCSEs, because a key part of the argument for creating Growth Fund was a concern, expressed in research, about small VCSEs’ lack of access to unsecured lending.

The cushion of Grant B facilitates this risk-tolerance arrangement; And any losses that can’t be covered by Grant B have to be taken on the chin by BSC as the wholesale investor - and hence a circa 5% compound interest rate that also comes with the BSC capital. The upshot in practical terms is that it’s the lenders — more than VCSE investees themselves — whose financial exposure is legally protected by the structure. And not all lenders however reported being happy with having to pass on BSC’s risk-mitigating 5% in their offer; their feedback indicates it can put drag onto their ability to clinch deals and thus the speed with which they can generate returns and cover costs when grant A runs out. (Note: Arrangements have changed since Covid).

The way Grant C has been used, has only served to reinforce the question about how the costs of risk are handed down (and how well grant B defrays this). Grant C has sometimes been deployed by lenders to discount the risk-bearing interests rates they’d otherwise have to pass on to VCSE investees. Lenders have also used Grant C in other inventive ways to mitigate risks for both them and their investees and themselves:

• Directly funding a social impact role or activity

• Covering a gap between a property mortgage loan offer and the purchase price

• Funding a necessary staff role for a new venture

• Cashflow funding for public sector contracts on investees’ books

• Funding upfront costs for opening or refurbishing premises

• As quasi-equity, with repayment required only if income surpasses a threshold

Grant C could be offered at up to 50% of each deal, at lenders’ discretion. Some lenders have preferred to use Grant C more than others. Two have not used it at all. One other lender ran their first Growth Fund scheme offering no Grant C at first, but, to develop its pipeline into a more diverse target audience for its second fund, found it needed it after all. 

Although the Growth Fund’s tally of 499 deals already done by June 2020 with 422 VCSEs shows that demand is real, lenders have reported one reason why pipelines have been slow to convert into loans arranged, is because VCSEs will typically opt for 100% grant if they can find it elsewhere on the way. In their view, loans can’t be made attractive to many VCSEs unless 20% or more of the deal is a grant. 

As Grant C came to lenders as a part of the three-in-one subsidy, they had to decide how balance grant C against their need for Grants A and B. Taking the subsidy pot as a whole across the lenders (£22.5m), Grant C is half the size of Grant B, but three times the size of Grant A. Most have offered Grant C sparingly at 15-30% of deal size – and could have done with more.

So what can we take away from this? 

Interestingly, the recommendations haven’t included increasing the amount of subsidy relative to principal capital available to the programme, which is 50/50 (and if they had, you’d have to ask probing questions about whether this is the best use of grant). Instead, the evaluators found that if key aspects of the programme’s design had been looser, some of the lenders that have struggled most could have worked their way out of a hole rather than sinking deeper into it. 

The learning suggests that relaxations made latterly (pre and during Covid19) helped when they were agreed – but with hindsight, it would have been better to have put some — if not all — of them into play from the start, alongside some other measures. So if you want to borrow ideas from the Growth Fund to subsidize future unsecured lending, we’ve learned from the lenders that the following recommendations could help you create a stronger footing:

• Grant A allocation needed to be higher than 10% of operating costs – especially for new lenders.

• Grant A (and the asset: liability ratio tests) also needed to be structured to cover a longer period of initial operational costs.

• And as part of this, additional Grant A could be made repayable, proportionate to revenue generated after a longer initial runway period.

• More centralised support is needed too, especially by newer lenders, to promote the loan schemes and help capture and convey the social impact achieved by them.

• Access to an agreed Grant B pot for a Fund should no longer be tied in proportion to each capital draw down. Instead it could be offered either as a discretionary first loss Fund, accessible throughout the Fund period or, entirely up-front (but handed back if not used) using a risk matrix to help guide — but not rule — newer lenders’ allocations of grant B.

• Less due diligence should be required of loans below £50k.

• Lenders could cover patient and growth capital needs better if they could apply different rules to recycled capital, e.g. offering follow-on loans/quasi-equity over £150k as ventures prove their worth.

• Technical support for new lenders could then help them move into offering these more sophisticated deal structures (quasi equity, use of SITR etc in supra £150k loan requests).

• Grant C could be used more flexibly at points throughout the loan period (not just at the start) – e.g. to help VCSEs mitigate difficulties, develop capabilities or seize opportunities.

Arguably, (although not suggested as part of the evaluation’s findings) it might also have been neater if Grant B had been deployed as the initial (recyclable) capital; in this way it could have emulated privately-sourced social investment but, with 0% interest charged to lenders while they established their pipelines, until they were ready to draw down larger volumes of capital that come with a price tag? 

Hindsight is a marvellous thing. And our final analysis on what we think would work better can’t be fully formed until we’ve taken stock when the programme ends. We started out on this experiment using some best guesses. The flex we’ve applied as the picture of reality became clearer will help to develop and sustain our partnership’s achievement of the fastest-growing part of the Social Investment market, while successor schemes emerge; all this hindsight is rich pickings for foundations considering backing future small-scale lending.

And aside from the role of BSC’s capital in the mix, for me there’s a sub plot in the story where perhaps where the most interesting lessons are yet to be found. A handful of our new lenders, are Community Foundations who instead of using BSC money, found alternative capital sources and had a variety of arrangements for covering operating costs too. One of them only sought a third of the Grant B required by other funds. Their relationships with Local Authorities and their own funding strategies are well worth you taking a closer look.
We will be.