New Interest-Rate Statistics
Tina Christoffersen and Marie Jakobsen, Statistics
Introduction
As from January 2003 new and significantly improved interest-rate statistics have been introduced for monetary financial institutions (MFIs)[1] in Denmark.
The statistics are based on a regulation from the European Central Bank, ECB, defining the framework for pan-European interest-rate statistics. The regulation is part of the implementation of the EMI's statistics package from 1996[2] on the need for harmonised statistics for the ECB's monetary-policy analyses.[3]
Denmark has a tradition for compiling statistics in accordance with international standards. Consequently, Danmarks Nationalbank decided to introduce the new interest-rate statistics at the same time as other European countries. On designing the statistics Danmarks Nationalbank has also attached importance to limiting the burden on the respondents and ensuring the practical value of the statistics to the financial sector. The statistics have therefore been implemented in close cooperation with the sector, which was involved at an early stage. During the preparation of the regulation the Danish MFIs were consulted on methodological issues and issues relating to the costs of developing and operating the statistics.
This article first describes the interest-rate statistics previously compiled in Denmark. This is followed by a brief outline of the content of the pan-European interest-rate statistics. Finally, the new Danish interest-rate statistics are presented, with focus on the interpretation of the statistics.
Previous danish interst-rate statistics
Until the beginning of 2003, the compilation of interest-rate statistics from Danish MFIs only comprised banks.
From the 4th quarter of 1995 up to and including 2001, quarterly interest rates for banks were published on the basis of information reported to the Danish Financial Supervisory Authority (FSA). Interest rates were calculated as average interest rates, i.e. as the quarter's interest income (expenditure) as an annual percentage of the quarter's average lending (deposits).
In 2002 the Danish FSA's average-interest-rate statistics were replaced by the FISIM statistics[4] that are compiled by Danmarks Nationalbank and published by Statistics Denmark. The FISIM statistics can also be used to derive quarterly interest rates according to the average-interest-rate principle.
These interest rates are, however, not fully comparable with those from the Danish FSA's statistics. This is primarily due to the fact that the FISIM statistics are based on international standards defined in the European System of National Accounts (ESA 1995), while the Danish FSA's statistics were based on the Danish order on financial statements.[5] In addition, the reporting population differs since the FISIM statistics are based on reports from the 24 largest banks in Denmark, while 98 banks reported to the Danish FSA.
New pan-european interest-rate statistics
At the end of 2001 the Governing Council of the ECB adopted a regulation on new pan-European interest-rate statistics for MFIs to take effect on 1 January 2003.[6] The central element in the preparation of the regulation has been requirements relating to the ECB's monetary-policy analyses. In addition, much importance has been attached to the burden on the respondents. In this connection the national central banks have been actively involved in the preparatory work for the new statistics. This also applies to Danmarks Nationalbank, which – at working-group level – has influenced the final wording of the regulation.
The pan-European interest-rate statistics concern the MFIs' monthly interest rates on lending and deposits vis-à-vis households and non-financial corporations. The statistics are based on the standards defined in ESA 1995. The statistics comprise both existing lending and deposits (outstanding amounts) and new lending and deposit agreements concluded in a given month (new business). The statistics on outstanding amounts to a large extent follow the definitions in the existing balance-sheet statistics for MFIs. The concept of new business is, however, new in a statistical context. The statistics on new business are described in Box 1.
|
Statistics on new business
|
Box 1
|
|
Statistics on new business provide information on the average interest rates on the month's new lending and deposits, and reflect the current market conditions. Definition of new business Breakdown by instrument types Breakdown by maturity |
|
During the preparation of the regulation, the interest-rate concept has been discussed in particular due to the different structures and existing interest-rate statistics of the individual countries. As a result, the regulation operates with two possible approaches to statistics on outstanding amounts. One option is to collect interest rates from the reporting institutions. The rate of interest is hence calculated as a weighted average of the rates of interest applying to outstanding amounts at month-end, corresponding to the annualised agreed rate.[7] The other option is to apply the average-interest-rate principle.
In this connection Danmarks Nationalbank has argued in favour of applying the average-interest-rate principle in line with central banks in other countries where interest-rate formation to a high degree depends on the ongoing development in the money-market rate. This has been based on a conviction that the average-interest-rate principle gives the most true and fair view of the interest on loans and deposits at variable interest rates. Applying the average-interest-rate principle has also made it possible to maintain the method that was already well-established among the Danish banks.
New danish interest-rate statistics
The new Danish interest-rate statistics comprise the largest banks in Denmark, i.e. the banks reporting to the FISIM statistics, as well as all Danish mortgage-credit institutions and two other credit institutions.[8] The composition of the reporting population is partly based on the regulation, and partly on what is most appropriate from a national point of view.[9] In this connection importance is attached to exempting the small institutions from reporting requirements since their importance to the statistics would not match the reporting burden.
Owing to different product portfolios it has been most appropriate in the Danish implementation to state interest rates separately for the individual types of institutions, even though the ECB's regulation only takes the aggregate MFI sector into account. Below is a description of the implementation of the regulation for banks and mortgage-credit institutions, respectively.[10]
Interest-rate statistics for the banks
For the banks it has been essential that the new interest-rate statistics on outstanding amounts could be founded on the FISIM statistics, which are based on the same standards as the regulation. In this connection it has been essential to be able to maintain the average-interest-rate principle, which is described in Box 2.
|
The average-interest-rate principle
|
Box 2
|
|
Interest rates calculated according to the average-interest-rate principle indicate the actual interest per annum payable on loans and deposits in a given period. The interest rates are calculated implicitly as the banks' interest income (expenditure) in a given period as a ratio of the average lending (deposits) in that period. The calculation of the average interest rates takes into account that the Danish banks generally operate with quarterly interest accrual on loans, and annual interest accrual on deposits. Furthermore, Danmarks Nationalbank's calculation of monthly interest rates is based on the actual number of days in the month in question (calendar days). This is the most commonly used interest-rate convention for lending and deposits in Danish banks. The monthly average rate of interest on loans is calculated as follows: Ru(m) = the month's interest income from lending, U(m) = the month's day-to-day average outstanding loans, N(m) = lending at zero interest, month-end, m = number of days in the month, and q = number of days in the quarter. The monthly average rate of interest on deposits is calculated as follows:
Ri(m) = the month's interest expenditure on deposits, I(m) = the month's day-to-day average outstanding deposits, and m = number of days in the month. |
|
With the new interest-rate statistics, the FISIM statistics have been expanded with a more detailed breakdown by instrument type and maturity of lending and deposits vis-à-vis households and non-financial corporations. In addition, the frequency of compilation has been increased from quarterly to monthly. There are no data breaks in connection with the transition to the new interest-rate statistics, and the interest rates are thus fully comparable to the quarterly FISIM interest rates for 2002.
Tables 1 and 2 illustrate the interest rates broken down by sector for respectively lending and deposits back to the 1st quarter of 2002. For deposits from households and non-financial corporations the figures are also broken down as demand deposits and time deposits. This breakdown is part of the expansion introduced in January 2003.
| The banks' average interest rates – outstanding lending |
Table 1
|
||||||
|
2002
|
2003
|
||||||
| Effective rate of interest, per cent p.a. |
Q1
|
Q2
|
Q3
|
Q4
|
January
|
February
|
March
|
| Total lending |
6.4
|
6.3
|
6.5
|
6.3
|
6.0
|
6.0
|
5.7
|
| General government |
3.4
|
3.8
|
3.9
|
3.9
|
4.0
|
4.1
|
4.2
|
| Financial corporations |
3.6
|
3.6
|
3.8
|
3.7
|
3.7
|
3.4
|
3.2
|
| Non-financial corporations |
6.1
|
6.1
|
6.1
|
6.1
|
5.9
|
6.0
|
5.8
|
| Households |
9.2
|
9.1
|
9.2
|
9.0
|
8.3
|
8.4
|
8.2
|
| Note: Interest rates comprise domestic lending in Danish kroner. The drop in interest rates between the 4th quarter of 2002 and January 2003 reflects the lowering of the official interest rates in early December 2002, which was not fully reflected until January 2003. The rate of interest on general-government lending has, however, increased. This is a consequence of a change in the reporting population with effect from January 2003, which has raised the level of interest rates on lending to this sector by approximately 0.5 percentage points. Source: Danmarks Nationalbank. |
|||||||
| The banks' average interest rates – outstanding deposits |
Table 2
|
|||||||
|
2002
|
2003
|
|||||||
| Effective rate of interest, per cent p.a. |
Q1
|
Q2
|
Q3
|
Q4
|
January
|
February
|
March
|
|
| Total deposits |
2.6
|
2.5
|
2.6
|
2.5
|
2.3
|
2.3
|
2.2
|
|
| General government |
3.1
|
3.0
|
3.3
|
3.2
|
2.6
|
2.8
|
2.6
|
|
| Financial corporations |
3.4
|
3.1
|
3.2
|
3.1
|
2.9
|
2.7
|
2.6
|
|
| Non-financial corporations |
2.9
|
3.0
|
3.0
|
2.8
|
2.4
|
2.4
|
2.3
|
|
| - of which demand deposits |
…
|
…
|
…
|
…
|
2.1
|
2.2
|
2.1
|
|
| - of which time deposits |
…
|
…
|
…
|
…
|
2.9
|
2.8
|
2.7
|
|
| Households |
1.9
|
1.9
|
2.0
|
1.9
|
1.8
|
1.8
|
1.7
|
|
| - of which demand deposits |
…
|
…
|
…
|
…
|
1.5
|
1.5
|
1.5
|
|
| - of which time deposits |
…
|
…
|
…
|
…
|
2.2
|
2.2
|
2.1
|
|
| Note: Interest rates comprise domestic deposits in Danish kroner. The drop in interest rates between the 4th quarter of 2002 and January 2003 reflects the lowering of the official interest rates in early December 2002, which was not fully reflected until January 2003. Source: Danmarks Nationalbank. |
||||||||
The new aspect for the banks is the statistics on new business.[11] Table 3 shows an extract from the banks' new-business statistics for the first three months of 2003.
| The banks' average interst rates – new lending and deposits |
Table 3
|
||
|
2003
|
|||
| Effective rate of interest, per cent p.a. |
January
|
February
|
March
|
| Loans, excluding overdraft facilities |
|
|
|
| Non-financial corporations |
4.2
|
3.8
|
4.1
|
| Households |
8.2
|
8.3
|
8.0
|
| - housing purposes |
7.4
|
7.5
|
7.4
|
| - consumer credit |
9.9
|
9.3
|
9.3
|
| - other |
7.3
|
7.8
|
7.3
|
| Deposits (time deposits) |
|
|
|
| Non-financial corporations |
2.9
|
2.8
|
2.8
|
| Households |
2.5
|
2.5
|
2.5
|
| Note: Interest rates comprise domestic lending and deposits in Danish kroner. Source: Danmarks Nationalbank. |
|||
In the bank's interest-rate statistics the individual interest rates cover a wide range of business types and customer relations. The banks' interest rates thus depend on the individual business transaction whereby the scope of the transaction, the customers' credit standing and any collateral pledged may have a significant effect on the interest charged on a loan. In this connection it is important to be aware of the differences between the statistics on outstanding amounts and new business, respectively.
In the statistics on outstanding amounts, the transactions included in the statement will to a large extent overlap from month to month. The interest on outstanding amounts is thus primarily affected by changes in interest rates on existing loans and deposits, and to a lesser degree by changes in the customer and/or product mix.
In the statistics on new business, there is no overlap between the transactions included in the statement from month to month. The average interest rates for a given month reflect the actual lending and deposit agreements concluded, whereby the interest rate on each agreement – in addition to the market interest rate – reflects the specific transaction and customer relationship. This also means that the development in interest rates from month to month may be influenced by changes in the customer and/or product mix.
On comparing the interest rates for respectively outstanding amounts and new business it can be seen that the lending rates for households are almost identical in the two statements. The reason is that most Danish banks' lending to households comprises loans at variable interest rates.
The banks' lending to non-financial corporations is also typically at variable interest rates, but in this case the rate of interest on new lending is at a lower level than on outstanding loans. The low level of interest rates on new lending to non-financial corporations during the period is attributable to a large proportion of repo transactions.
On the deposit side, it is only possible to compare the rate of interest on outstanding and new time deposits, since other deposit types are not included in the new-business statistics, cf. Box 1. Part of the explanation for the similar rate of interest on outstanding and new time deposits for non-financial corporations is that a large proportion of their time deposits are special-term deposits at variable interest rates. For households, time deposits will typically be at a fixed rate of interest. In this connection it should be noted that the rate of interest is higher for new than for outstanding time deposits. This contrasts with what might be expected in the light of the falling level of interest rates. One reason is that products at a very low rate of interest (e.g. premium savings accounts) constitute a significant share of the outstanding time deposits, but only a small proportion of the new time deposits.
Interest-rate statistics for the mortgage-credit institutions
The mortgage-credit institutions have not previously reported interest-rate statistics and hence these are built up from scratch.
The interest-rate statistics of the mortgage-credit institutions operate with an effective rate of interest that takes into account coupon rates, administration fees and discounts/premiums on issue of the underlying bonds. This is the most precise expression of the rate of interest payable on a mortgage-credit loan by the borrower.[12] It also complies with the ECB's interest-rate concept defined as the annualised agreed rate.
The effective rate of interest is primarily determined by the bond yield[13] at the time when the underlying bonds are issued. Only the fee, which is the mortgage-credit institutions' interest margin, is determined by the mortgage-credit institution and can thus be customer-specific.
Tables 4 and 5 show the average effective rates of interest on outstanding and new mortgage-credit loans in the first three months of 2003. For the mortgage-credit institutions the statistics on new business are the most interesting. This is because a large proportion of the outstanding loans are at fixed interest rates, and because the statistics on outstanding amounts do not distinguish between fixed-rate and adjustable-rate loans.[14]
| Mortgage-credit institutions' effective interest rates, including fees – outstanding loans |
Table 4
|
||
|
2003
|
|||
| Effective rate of interest, per cent p.a. |
January
|
February
|
March
|
| Households |
6.6
|
6.5
|
6.4
|
| Non-financial corporations |
6.4
|
6.4
|
6.3
|
| Note: The interest rates comprise domestic lending in Danish kroner. Source: Danmarks Nationalbank. |
|||
| Mortgage-credit institutions' effective rate of interest, including fees – new lending[15] by fixed-interest-rate period |
Table 5
|
||
|
2003
|
|||
| Effective rate of interest, per cent p.a. |
January
|
February
|
March
|
| Households |
5.3
|
5.3
|
5.3
|
| £ 1 year |
3.7
|
3.6
|
3.5
|
| 1-5 years |
4.4
|
4.1
|
4.1
|
| >5 years |
5.9
|
5.8
|
5.8
|
| Non-financial corporations |
5.0
|
4.9
|
5.3
|
| £ 1 year |
3.3
|
3.1
|
3.2
|
| 1-5 years |
4.2
|
3.9
|
4.1
|
| >5 years |
5.7
|
5.5
|
5.5
|
| Note: The interest rates comprise domestic lending in Danish kroner. Source: Danmarks Nationalbank. |
|||
In the interest-rate statistics for the mortgage-credit institutions the individual rates of interest reflect loans at different maturities of the underlying bonds. For the mortgage-credit institutions, the lending rates of the statistics are thus affected by the maturity structure. This is supported by the development in interest rates shown in Table 5, where the rate of interest on total new lending to households remained unchanged in the period, while the rates of interest within the individual maturity bands were declining.
Generally, the interest rates in the period are higher for households than for non-financial corporations, even though all mortgage-credit loans are based on the same bond series. One explanation is the fee, which is typically higher for households than for non-financial corporations.[16] Another explanation may be that fixed price agreements and price cut-offs, which entail a higher effective rate of interest, are more prevalent for households than for non-financial corporations. In addition, there may be differences in the average term to maturity of the underlying bonds, cf. above. Finally more active debt management by non-financial corporations may explain why their interest rates on outstanding loans are lower than those of the households.
Conclusion
The new interest-rate statistics constitute a significant improvement of Danish interest-rate statistics. Firstly, the banks' interest-rate statistics on outstanding amounts have been expanded and changed from quarterly to monthly statements. The increased frequency makes it possible to monitor the development in the banks' average interest rates more closely. In this connection the new-business statistics constitute a relevant supplement. Secondly, interest-rate statistics have now been established for the mortgage-credit institutions, which are a major lender in the Danish market.
The interest-rate statistics can be used as e.g. input to analyses of the monetary-policy transmission and other economic analyses, as well as analyses of the development in the financial sector. The latter may be analyses of circumstances relating to competition and profitability. Finally, the interest-rate statistics may give the MFIs information on their own competitiveness.
The interest-rate statistics can be used in connection with the other MFI statistics, which are fundamentally founded on the same definitions. Structuring the interest-rate statistics according to the guidelines in the ECB's regulation also facilitates comparisons with other European countries.
The new interest-rate statistics were first published on 30 April 2003 with interest rates for the first three months of 2003. In future, interest-rate statistics will be published on the 19th banking day of each month.
[1] Monetary Financial Institutions in Denmark comprise banks and mortgage-credit institutions, other credit institutions that are neither banks nor mortgage-credit institutions, money-market funds and Danmarks Nationalbank. Danmarks Nationalbank and money-market funds are not covered by the interest-rate statistics, however.
[2] See Kim Abildgren, Kommende EU-statistikker på det finansielle område (Future EU statistics in the financial area – in Danish only), Danmarks Nationalbank, Monetary Review, August 1996.
[3] The first step in the implementation of the statistics package comprised harmonised balance-sheet statistics for MFIs. These were implemented in Denmark with effect from July 2000. For a detailed description of the balance-sheet statistics, see Kjeld Ole Nielsen and Anders Mølgaard Pedersen, New Balance-Sheet Statistics for Monetary Financial Institutions, Danmark Nationalbank, Monetary Review, 4th Quarter 1999. In connection with the introduction of the new interest-rate statistics the balance-sheet statistics have also been expanded, primarily with harmonised flow statistics.
[4] FISIM is an abbreviation of Financial Intermediation Services Indirectly Measured. FISIM statistics are based on requirements from Eurostat and are aimed at supplementing the compilation of the gross national product with information on the part of the banks' income which does not relate to commission and fees, etc., but is a result of the interest margin. Statistics Denmark is responsible for the methods applied to the FISIM statistics.
[5] The data break primarily relates to sector definitions.
[6] Regulation (EC) No. 63/2002 of the European Central Bank of 20 December 2001 concerning statistics on interest rates applied by monetary financial institutions to deposits and loans vis-à-vis households and non-financial corporations (ECB/2001/18).
[7] The annualised agreed rate, r, is defined as: r = (1+ra/n)n – 1, where ra is the interest rate per annum agreed with the household or non-financial corporation, and n is the number of times that interest is accrued per annum.
[8] KommuneKredit and Danish Ship Finance.
[9] The reporting population corresponds to the full-reporting institutions in the balance-sheet statistics.
[10] In the implementation, the two other credit institutions have been treated together with the mortgage-credit institutions. They are not included in the national publication, since interest rates are published separately by type of institution. With only two other credit institutions in the reporting population, publication is not possible for confidentiality reasons. No further reference is made in this article to the interest-rate statistics for other credit institutions.
[11] The regulation's definitions in the statistics on new business, cf. Box 1, can be immediately transferred to the banks' product portfolio. Unlike for the statistics on outstanding amounts, the banks report interest rates to the new-business statistics. This is a requirement under the regulation. The interest rates are, however, calculated using the average-interest-rate principle. Along with the interest rates the volume of new business is reported. The business volumes of the individual banks are used to co-weigh the interest rates across the institutions.
[12] The calculation of the effective rate of interest is based on an assumption that the loan is held until maturity. The interest-rate concept thus does not take the call option into account, which means that typically the rate of interest calculated when the loan is established will not reflect the rate of interest over the entire term of the loan for callable loans.
[13] The bond yield is determined by the coupon rate and any discount/premium on issue.
[14] The fact that the outstanding volume of loans is dominated by fixed-rate loans is in accordance with the interest rates in Tables 4 and 5. The interest rates on outstanding loans are thus at a higher level than for new business. However, conversion activity is reflected in the marginally declining rates of interest on outstanding loans, cf. Table 4.
[15] For the mortgage-credit institutions, the regulation's definition of new business, cf. Box 1, has been translated as follows: Newly-established loan agreements, conversions of existing loans to another coupon rate, maturity or currency, conversions between fixed-rate and adjustable-rate loans, and changes in the rate-adjustment profile (i.e. the frequency of interest-rate adjustment and the proportion of the loan that is subject to adjustment). Adjustment of the rate of interest on an adjustable-rate loan is not new business.
[16] The statement for the first three months of 2003 shows a difference in the average fee for households and non-financial corporations of approximately 0.1 percentage point for both outstanding and new lending.