4 A STATISTICAL OVERVIEW OF UK CREDIT UNIONS
4.1 Obtaining data on UK credit unions is difficult. The FSA does not make data available on individual credit unions although up to 2001 this data was provided by the then Regulator the Registrar of Friendly Societies 4. In Northern Ireland ( NI) the Registrar still provides data to researchers on a credit union by credit union basis. This section is based upon: 1) the annual return by the FSA which provides some data on credit union by region; 2 aggregate quarterly returns made available by the FSA for the movement as a whole in GB; 3) the researchers' own database on Irish credit unions in NI; and 4) a survey which the researchers undertook in 2006 which enabled them to collect 112 of the annual reports of Scottish credit unions in Scotland.
4.2 In Table 3 (of the Statistical Annex) summary information is present on credit unions between 2002 and 2006 for each of the four regions. It is clear that relative to the size of the respective regions the movement in NI is strongest while the movement in Wales is quite weak. From Table 3 it can also be seen that while credit union numbers are declining, although not in NI, membership is increasing in each region as are loans extended to members.
4.3 In Table 4 (of the Statistical Annex) we use FSA quarterly data to summarise on a six month basis from June 2003 to June 2007 (September 2007 data is available and we have added it for completeness sake) credit union numbers, members, total loans extended and average loans per member. This information is only provided for GB credit unions and is based upon those making returns to the FSA. Table 4 reveals a pronounced decline in credit union numbers. This has occurred, in the main, through credit union transfer of engagements, although approximately 20 credit unions have failed over the period with members compensated through the FSA's compensation scheme. Membership has increased as has total loans, however, when we compute the ratio loans per member, which we do not deflate, we notice that there is a slight downward trend in this nominal ratio suggesting that, in real terms, there has been a decline in the average loan being extended to credit union members.
4.4 In Table 5 (of the Statistical Annex) we use FSA quarterly data to summarise on a six month basis from June 2003 to June 2007 (plus September 2007 data) income and expenditure with the ratio expenditure/income also calculated. We note that this efficiency ratio has trended upwards over the period and is now touching almost 80%. The Irish League of Credit Unions suggests that for its member credit unions in Ireland that this ratio should be less than 43%. The analysis in Table 5 reveals credit unions in GB to be subject to costs which are too high for the income generated by the credit union.
4.5 In Table 6 (of the Statistical Annex) we use FSA quarterly data to summarise on a six month basis from June 2003 to June 2007 (plus September 2007 data) bad debt write-offs and bad debt provisions we also compute the ratios write-off per member and provisions per member. This data emphasises that bad debt is increasingly becoming a problem for credit unions in GB. In September 2007 the bad debt write-off per member was £3.75 compared to £0.90 in June 2003. Provisions per member have also steeply increased over the period.
4.6 As indicated the FSA does not make available data on individual credit unions. In Table 7 and Table 8 (of the Statistical Annex) we present a profile of 112 Scottish credit unions for the period end 2005. This information is based on a 2006 survey by the researchers. Note at that stage there were 134 credit unions in Scotland so the response rate was high. In each table we analyse credit unions by member size bands - over 5,000 members, 2,000 to 5,000 members, 1,000 to 2,000 members, 500 to 1,000 members and less than 500 members.
4.7 In Table 7 we detail credit union numbers, members, loans and loans per member. We note that there is a pronounced size effect with larger credit union making higher value loans to their members. In Table 8 we present income, expenditure and the ratio expenditure/income. We note that both small and large credit unions in Scotland have the same expenditure to income ratio. This suggests little evidence of economies of scale. In Table 9 (of the Statistical Annex) we detail bad debt write-offs and bad debt provisions and also compute the ratios write-off per member and provisions per member. Again we do not note any size effect in either write-offs per member or provision per member. However, we do observe that bad debt write-offs and provisions are in general quite high across all credit union size categories in Scotland.