The relationship between religion and economic behaviour has long proved a fascinating and rewarding field of study. Divine worship and the struggle for material betterment are perhaps the two most powerful, and frequently conflicting, drivers in human affairs.
But because a subject is worthy of study does not mean that the research involved will be easy or even straightforward. A number of questions need to be resolved before work can begin. Questions such as: What will be our population group? What religious event or events will form the basis of our observation?
For our study we chose the Islamic faith, the month of Ramadan and the population of Turkey. In terms of the economic behaviour that formed the basis of our research, it was the granting of small-business loans. We wanted to know whether loans advanced during Ramadan behaved differently in terms of outcomes than those granted outside the month.
What we discovered was that they did so markedly, with significantly higher delinquency rates for loans made during Ramadan as against the rest of the year.
The scope of the study
For our work to produce meaningful results, we needed a society that was, relatively speaking, both economically advanced and religiously homogenous. A 2012 survey found 97% of Turks identified as Muslim and 87% said they observed the injunction to fast in Ramadan during daylight hours. Such overwhelming identification with one faith is unique in a member of the Group of 20 leading economies.
We then required a class of lending in which the personal decisions of bank officers, acting alone, would be the key factor. Personal loans would not qualify in this regard, as those lending decisions are now largely automated. At the other end of the scale, loans to medium-sized and large business enterprises are made by credit committees, in which the individual loan officer is one among many, greatly reducing the scope for individual misjudgement to affect the outcome.
Lending to small businesses, by contrast, is an area in which bank officers enjoy a wide measure of discretion, not least because of the lack of reliable financial statements in this sector and the widespread existence of undocumented assets and sources of income. The lending officer’s judgement is thus of critical importance.
We defined small-business loans as those in which the capital sum amounted to one million Turkish liras or less, in 2008 values, broadly equivalent to US$770,000 (£552,000). Non-cash loans were excluded, as was lending tied to the value of a precious metal and loans where record-keeping was unsatisfactory for whatever reason. Finally, we excluded loans made by State-owned banks, because of the potential for non-commercial, politically influenced decision-making by such institutions.
Studying official data of all small-business loans made during the period 2003-2013, we found that those advanced during Ramadan are between 10% and 15% more likely to become delinquent within two years than those granted outside Ramadan. Put another way, the average default rate on loans originated during Ramadan is 2.68%, 38 basis points higher than the average rate for loans granted during the rest of the year, 2.3%.
Why might this be? One obvious suggestion arises from a feature of Ramadan on which we have already touched – fasting. Muslims are required during the month to abstain from food and drink from dawn until sunset. More than two-thirds of adults in Turkey observe the fast. Could the effects of this self-denial have an impact on the quality of lending decisions?
Fasting and finance
We find this is indeed the case. Turkey contains two types of banks: conventional, western-style institutions and so-called participation banks, usually known as Islamic banks, which uphold Islamic teaching on financial practices, not least the prohibition of usury. Rather than lend money at interest, such banks take stakes in their customers’ businesses, hence the “participation” label.
It is not unreasonable to assume that loan officers in the latter type of bank are more likely than those in the former to be observing the fast, and the “Ramadan effect” on loan quality is indeed more pronounced in the participation banks than in conventional institutions.
Furthermore, the Ramadan effect was more pronounced on days when observing the fast was more challenging than normal. Ramadan is not a fixed period in the calendar but moves earlier by about 11 days a year as measured by the western Gregorian calendar, completing a cycle of all the seasons about once every 33 years.
Medical studies have showed that observing the Ramadan fast at any time of year is associated with moderate increases in dehydration, along with headaches, sleep deprivation and feelings of exhaustion. When the feast falls during the summer months, or when some days are markedly hotter than others, these side-effects are likely to be magnified.
When Ramadan falls in the autumn, survey evidence has shown, about 70% of Turkish professionals observe the fast, but this drops to nearer 50% when Ramadan falls in the summer. Our research shows the quality of lending is lower on these more challenging fast days than on others.
Charity plays a part
A second explanation, complementary rather than in conflict with the first, relates to a second injunction to be observed by the faithful during Ramadan – the giving of alms, or charitable gifts. Our research found that the Ramadan effect is more pronounced when financially strong lenders are advancing loans to financially weak borrowers. This suggests that the charitable impulse encouraged during Ramadan influences, consciously or not, supposedly commercial decision-making.
One of the most remarkable aspects of the Ramadan effect is that, despite their higher delinquency rates, Ramadan loans have lower credit spreads than those originated during the rest of the year, 4.38% on average against 4.85%. This despite the fact that Ramadan and non-Ramadan loans involve similar risk, collateral levels and maturity dates.
We have scrutinised various alternative explanations. One would be that the type of small-business loan applicant in Ramadan differs from those seen during the rest of the year. But the absence of a similar Ramadan effect on personal loans, where decision-making is largely automated, would seem to rule this out, while the lower credit spreads point to loan-officer misjudgement as a much more likely explanation than the appearance during Ramadan of a significantly different population of borrowers.
Similarly, our focus on small-business loans – where human judgement is central – helps to underline the scale of misjudgement during the month.
Nor is the Ramadan effect a parochial issue, of interest only to Muslim countries, even ones as economically advanced as Turkey. In 2010, it was estimated that there were 1.6 billion Muslims in the world, just under a quarter of the global population. That means that managerial errors committed during Ramadan have the potential to generate significant worldwide effects through the transmission mechanisms of trade and investment.
Our research provides, for the first time, we believe, evidence that a religious practice has a material influence on economic decision-making. It points to two possible causes, both of which can be present in the same person, for loan-officer misjudgement: spiritual, as seen in the quasi-charitable advancing of loans by stronger lenders to weaker borrowers, and physiological, shown by the greater Ramadan effect seen in loans made during warmer summer days.
Although at this point we can’t conclude whether Ramadan leads to worse overall allocation of credit, for example, it is possible that the net effect on welfare is positive if Ramadan
fasting strengthens self-control and improves credit-market outcomes outside of Ramadan, we hope our work will encourage much greater research into this subject.