Like many DFIs, BII has a fair share of its portfolio invested in financial services (currently 34 per cent of the portfolio and 26 per cent of commitments made in 2021). Some observers think we should invest less in financial services and more in sectors such as agriculture and manufacturing.
But investing in financial services is not an alternative to investing in agriculture and manufacturing so much as another way of going about it. We invest via financial institutions to get money into the hands of smaller businesses, some of which will be operating in sectors such as agriculture and manufacturing. [1] BII can make a relatively small number of high value investments directly, but to reach hundreds of thousands of small businesses in Africa and Asia, and millions of microenterprises, we would need to be a global bank with hundreds of branches and thousands of employees. [2]
Not every business that borrows from a bank or other lender will be operating in the sectors we would prioritise if investing directly, however. Some will be urban consumer services, for example. Which brings us to two important concepts: scale, and stages of market development.
When she was chief economist of the Foreign, Commonwealth & Development Office (FCDO), Rachel Glennerster would often emphasise the importance of scale (here is one example). Good development interventions are not small but perfectly formed — they do something positive for as many people as possible. There can be a trade-off between targeting support at specific areas, and scale. If you think some small businesses are more developmental than others, the most effective way to reach them at scale might be to help banks make credit broadly available to small businesses, and not worry that some of the businesses will be of the less developmental sort (and even they contribute something to poverty reduction by increasing demand for labour).
The same can be true when it comes to different types of activity within banks. Banks can reach many thousands of businesses, but they also do other things that are less important for development too (buying government bonds, offering credit cards and mortgages). In some circumstances the most effective intervention to get more of the developmental thing you want might be to encourage the growth of large banks that can deploy credit to firms at scale, and you accept getting more of everything else too. In other circumstances, it may be better to take a more targeted approach.
Some instruments are more targeted than others. Banks mostly finance their lending by borrowing, often from retail depositors, but capital adequacy regulations also require them to have a cushion of equity to absorb losses. [3] Investors can inject fresh equity, which banks can then leverage by borrowing. If we put £10 million of equity into a bank that could be leveraged to finance £50 million of lending and other activities. If 20 per cent of the bank’s assets are loans to businesses, that would amount to £10 million of new business lending. So, depending on things like leverage ratios and which activities the bank ends up growing fastest, £10 million of core support to bank, which cannot be targeted at particular activities, may result in more new lending to businesses than a dedicated £10 million credit line earmarked for business lending.
But these decisions are about more than the mechanics of different instruments. The best approach also depends on the stage of market development. In some circumstances, such as after the 1990s financial sector liberalisation in India, the priority may be to build stronger and more competitive banks and support their regional expansion. [4] We would like the banks we support to expand small business lending faster than other activities, but even if it shrinks in relative terms (business lending falls as a share of total assets) it is the absolute increase that matters.
When financial service providers (FSPs) are less well developed it also make sense to be less prescriptive, to maximise the chances that the activity you are trying to encourage will be commercial sustainability. When a bank is reluctant to grow lending to SMEs, the first hurdle is to demonstrate that can be done profitably. Imposing more demanding requirements makes things harder for lenders, so it should only be attempted when likely to succeed. [5]
Our objective is to support the development of vibrant local financial sectors that provide a range of products on competitive terms. Once generalist FSPs are well-established, attention can shift to pushing them further into specific areas, such as green lending or lending to female-owned businesses. Or we can support more specialist lenders.
Finance and poverty
The link between finance and poverty reduction also sometimes eludes observers of development finance. In rich countries, large financial sectors engage in activities of dubious social value (zero-sum speculation on financial instruments, extracting fees for wealth management, financing leveraged buy-outs etc.) and the evidence does suggest that beyond a certain point, finance becomes unhelpful. The economies of Africa and Asia are nowhere near that point.
There is a great deal of research into the links between finance and growth (here is one survey). The supply of finance also affects job creation. [6] One of the most persistent fallacies in some development commentary is that only inclusive growth counts towards poverty reduction. This is not true. Of course, growth accompanied by inequality reduction is more powerful, but even when accompanied by widening inequality growth can still increase the incomes of people living in poverty. Historically, growth as measured by the change in national average income accounts for 90 per cent of the variation in extreme poverty rates. [7]
The best approach to the financial sector will change with context. BII’s financial sector strategy addresses two dimensions of underdeveloped financial systems: strengthening the system itself (its efficiency, stability, and diversity) and its inclusivity (extending access to underserved customer segments and marginalised groups). We assess countries by their stages of market development and try to tailor interventions to match needs. To further those objectives, we focus on developing capital markets to help FSPs raise longer-term finance (a lack of which inhibits long-term lending); encouraging lending to SMEs within generalist FSPs and via specialist lenders; the provision of financial services as a basic need (credit, savings, insurance); harnessing digital financial services to reduce costs and promote inclusion.
The support that FSPs need from DFIs also varies with context, which implies that investments should be assessed based on market conditions at the time they were made. In some instances, banks that have not yet listed on the stock market struggle to raise equity from commercial investors. Elsewhere even listed banks may not be able to find longer-term debt financing in volume or may need risk-sharing facilities to expand into more impactful areas. In more mature markets, it may only be innovative specialist lenders that cannot raise capital from commercial investors.
The evidence on the impact of financial sector development on poverty is vast and sometimes contradictory. But the idea that a higher level of overall financial development will result in more equitable economic growth is supported by empirical evidence (Proano et al. 2023). The research listed in footnote 6 shows that even in middle-income countries, where the financial sector will be relatively better developed, expanding access to banking still has an impact. There are also likely to be complementarities between a stronger local financial sector and other development interventions, such as vocational training and policy reforms to lower barriers to firm entry. Openness to trade also tends to have broader-based benefits in countries with more developed financial sectors.
DFIs should make investments based on expected impact, knowing that things will not always turn out as hoped. My view is that although the strength of the relationship will vary over time and place, the importance of finance for growth and poverty reduction is beyond doubt, and that growth is more likely to be inclusive when the financial sector is too. That means expanding geographical coverage and extending finance beyond well-established firms that can post collateral is especially important. An independent evaluation that took stock of our financial sector portfolio was published in 2020. Our updated management response has been published here.
[1] India’s manufacturing sector is dominated by small firms, for example.
[2] This blog is about banks and other financial institutions that serve businesses; the evidence concerning microfinance institutions is summarised here.
[3] This is a simplification – there are other risk-bearing forms of capital.
[4] Financial liberalisation in India improved capital allocation, resulting in more investment, higher productivity, and wages (Bau and Matray, 2023). George et al. (2020) found the association between firm productivity and lending is stronger for private banks than public in India. Xu and MacDonald (2022) show that the ratio of credit to GDP had been below trend in India since 2012 (a “credit gap”) and that better capitalised banks in India are associated with higher credit and GDP growth. But later there was a real-estate led credit bubble that burst in 2019 (Subramanian and Felman, 2019).
[5] The same holds for private equity funds, where we will often start by supporting generalist fund managers in frontier markets and then shift our support towards more specialist impact funds once the local private equity ecosystem is better established.
[6] Because increased lending is both a cause and a consequence of growth it is hard to empirically isolate the impact of an increase in the supply of finance on growth. One way is to exploit the quasi-random arrival of banks in regions that lacked them. Kendall (2012) shows how the regional presence of banks affects growth in India; Fafchamps and Schundeln (2013) show how local banks resulted in faster firm growth in Morocco; Ji et al. (2022) show how bank expansion affected regional GDP in Thailand; Chakraborty et al. (2022) uses branch expansion in India to show how liquidity propagates through production networks, increasing sales and employment; Bruhn and Love (2014) show how bank branches affect the income of lower income individuals in Mexico.
[7] The relationship between finance and inequality is complicated – it depends on what type of finance for who in what circumstances (Eichengreen et al. 2021) – but stronger firms and better-off households may often be more able to take advantage of the opportunities afforded by improved access to finance than others. Financial liberalisation episodes are sometimes accompanied by an increase in inequality (Eklou and Foster, 2023). Some of the channels through which finance increases inequality tend to be active immediately, while the inequality-decreasing channels tend to operate with a lag (Avdjiev and Spasova, 2022).
