In turn, the participating organizations will take the money lent to them and pass it on to the consumers. So again, in the case of Bangladesh, Grameen Shakti and BRAC Foundation offered customers loans for up to three years at 12-13 per cent interest. It is in the difference between the refinance amount of 6 per cent and the retail interest rate of, for example, 12-13 per cent that these finance organizations make their money.

In Sri Lanka, it was much the same process. The administering unit (AU) worked through the DFCC bank, which passed on funds at roughly 11-12 per cent. Participating finance entities such as SEEDS then on-lent the money at higher rates of roughly 24 per cent. With a healthy spread on its loans, and with sufficient funds in place under the World Bank's ESD and subsequent RERED projects, SEEDS was able to radically scale up their financing of solar between 1999 and the end of 2007, by which time they had lent for 69,599 solar systems.21

SEEDS then set an example to the rest of its industry. In 2004, SEEDS was executing 1200 solar loans per month on average, with 90 per cent market share. But then other finance institutions entered, such as Lanka Oryx Leasing Company (LOLC), Ceylinco Leasing, Alliance Finance and Sanasa. By 2005, SEEDS's market share had reduced to 50 per cent. But even then it was doing more solar loans per month than all the other finance institutions combined.

—SEEDS Other PCIs Total loans

Figure 7.9 SEEDS concedes market share in Sri Lanka to other PCIs

Source: Finucane (2005), p19

It is important for policymakers to note, however, that if not executed well, lines of credit can get blocked. This comes down to the selection of the agency responsible for disbursing the funds and the criteria they adopt for disbursement. By and large, it is best if the agency is independent (for example not part of a government ministry or owned by the government), such as the DFCC Bank in Sri Lanka, which housed the AU. It is also important that a policymaker scrutinize and decide upon the criteria that will be used in assessing the participating finance institutions. If these are too rigid and conservative, they will limit the flow of funds and overall diffusion. As we saw in the case of the World Bank loan to India, because IREDA was not predisposed to lending to rural finance institutions for solar, and because their procedures and criteria were quite restrictive, the money never reached the rural markets. Instead it went to the commercial sector, which had a perceived lower risk of default and could better meet the paperwork and security requirements.

Having established a viable line of credit to participating finance institutions, how important is the rate of interest that customers bear? Should this be lower than the prevailing rates to incentivize diffusion?

The rates in Sri Lanka, at 24 per cent, were not low. They were even higher in the case of the Indonesian entrepreneur, at 30 per cent. But as the latter reasoned, it was possible to charge higher rates of interest provided the actual monthly amount the customer paid was no more than roughly US$10. To achieve this target figure, he offered his loans over a longer period of time (four years) than the existing rural banks in Indonesia were offering (just two years). The same was achieved by SEEDS, which initially offered their loans for five years. In both cases, with longer-tenure loans, we saw very high rates of diffusion. This is a key lesson for policymakers trying to encourage more finance for solar: do not focus on the interest rate, focus more on the duration of the loans. Finance institutions can charge a high rate of interest (to cover their costs and their risks) provided they can offer longer-term loans.

To further reinforce the message that high rates of interest do not necessarily restrict a rural solar market, we can point to a study from Kenya which found that the interest rate was more important to those who already had access to finance than to those who did not. Prospective customers who were part of a cooperative, and used to getting loans at 12 per cent, did not want to pay more than 12-14 per cent over two years for solar. Those in a very different setting, where there were no other formal credit routes, were willing to take on loans at 20 per cent over two years, which starts to approach the commercial rates in Kenya of 25-30 per cent.22

On the other hand, as discussed in Chapter 5, the case in India was rather different. In 2003 UNEP and the Government of India introduced a scheme whereby households in the state of Karnataka were able to buy a system at a reduced rate of interest, reduced from 12 per cent to just 5 per cent per annum, from two public-sector rural banks with extensive branch networks -Syndicate Bank and Canara Bank. This rate was subsequently gradually increased to 7 per cent, then to 9 per cent, and finally it returned to its original 12 per cent.

In this case, however, there was already a widespread network of rural banks engaged in financing solar at 11-12 per cent interest, and there were already several firms actively selling solar through this banking network. In this situation, it is not clear that reducing the interest rate had a terribly big effect on affordability. But it did have a psychological impact for the customer, giving the feeling of getting a better deal. Customers had already become familiar with offers from companies and participating finance institutions of 12 per cent - so when the opportunity came up to buy at 5 per cent, it created a stimulus that helped to accelerate diffusion, until, that is, the funds ran out.

So the conclusion must be that high rates of interest are not necessarily a deterrent, provided the duration of the loan is sufficiently long to compensate. And it is better to have high rates of interest than no finance at all for solar. That said, if there is an existing market for solar, and a well-established rate of interest that customers know, then a useful stimulus can be a low-interest loan scheme to excite demand. But it is important that, if such a stimulus is used, there be sufficient funds to sustain the scheme over a long period of time, for example 10 years, with interest rates gradually approaching market rates over that period.

Finally, what of those cases where no rural banking infrastructure exists? There are countries where the rural banking infrastructure is extremely weak, and where there just might not be any banking partners of MFIs to encourage to enter the market. Indeed, it is because some analysts have been faced with these situations - a total absence of rural lending - that they conclude:

The prospects for consumer credit are very specific to cultural, legal and financial factors in each country. The Sri Lanka micro-credit model appears sustainable, but perhaps only because Sri Lanka has a strong and long-standing microfinance culture and set of institutions in rural areas, along with a well-developed commercial banking system.23

Given the strength and reach of the microfinance industry throughout the emerging markets, it is hard today to conceive of a country that does not have any rural finance in place. But, assuming it's simply not possible to use a grant and lines of credit to encourage the entry of any finance players into the rural solar market, what should a policymaker do?

In some countries, entrepreneurs and policymakers have tried what is called 'fee-for-service' as a way of financing solar and ensuring customers had access to an ongoing service. Fee-for-service describes a situation in which a business retains the solar system on its own balance sheet as an asset, and then installs it in the home of a rural customer, charges a monthly fee, provides free replacement of parts, and hopes that the customer will pay on time and that their asset will not disappear. It is essentially a 'rental' model. Because it was deemed to have so much potential when launched at the end of the 1990s, the following section takes some time to review why it has generally failed. Specifically, I refer to the high-profile example of South Africa to substantiate why it is better to avoid a fee-for-service model.

0 0

Post a comment