Technical Note on the Tiered Mandatory Reserves System

CENTRAL BANK OF MADAGASCAR

MONETARY POLICY IMPLEMENTATION DIRECTORATE

TECHNICAL NOTE ON THE TIERED MANDATORY RESERVES SYSTEM

1. Objectives This note aims to explain the bank rating system intended for the application of Mandatory Reserves (MR) coefficients by tier. This mechanism allows the Central Bank of Madagascar (BFM) to optimize the structural liquidity management of the banking system and to encourage the financing of investments.

2. Calculation Method The indicators used to evaluate institutions subject to MR are calculated based on the average of available observed data over the last three rolling months.

The MR coefficients are subject to quarterly review by BFM.

The indicators used to construct the index for each bank are presented in the table below:

Table 1: Indicators and Measures with Scoring Rules

$I_i$	Indicators	Measures	Scoring
$I_1$	Interest Rate Spread	Difference between the lending rate and the deposit rate	9 if less than 3.00% <br> 0 if greater than 7.00%
$I_2$	Transformation Intensity	Total credit granted by the bank relative to the MR calculation base (Demand Deposits and Time Deposits) in Ariary	9 if greater than 95.00% <br> 0 if less than 85.00%
$I_3$	Credit Allocation	Medium and long-term credit relative to total credit	9 if greater than 60.00% <br> 0 if less than 50.00%
$I_4$	Mobilization of Liquidity Surplus	Share of untransformed liquidity surplus	9 if less than -5.00% <br> 0 if greater than 0.00%

The data used to calculate the scores for each indicator are those from month m-2.

---

Data Sources:

• For $I_1$, the data used for calculation are derived from the monthly declarations of rates made by each bank;
• For $I_2$, $I_3$, and $I_4$, the calculation elements are extracted from the monthly statements of the banks.

The score of a bank for an indicator $I_i^1$ depends on the position of its value relative to the minimum and maximum bounds set for that indicator. To ensure the evaluation is consistent and comparable across all banks, a linear interpolation calculation method is used.

The calculation methods for the score differ depending on the interpretation direction of the indicators.

• For indicators $I_2$ and $I_3$, where a higher value indicates better performance, the score is determined by a relationship increasing with the indicators;
• Conversely, for indicators $I_1$ and $I_4$, where a lower value is preferable, the formula is adjusted to reflect a decreasing relationship between the indicator and the score.

Thus, in all cases, the score evolves between the fixed bounds (from 0 to 9); an increase reflects an improvement in bank performance, regardless of the direction of variation of the considered indicator.

These various scores will then be used to calculate the global score for a bank according to the following formula:

$$Global Score of a bank = \sum_{i=1}^{4} W_i * Score_i$$

Where: $W_i$ represents the weight assigned to the score obtained from indicator $I_i$; this weight determines the relative importance of this indicator and reflects the strategic orientation of BFM's objectives (See Table 2). $Score_i$ denotes the score obtained by the bank relative to indicator $I_i$.

Table 2: Weighting of Each Indicator

Indicator ($I_i$)	Interest Rate Spread	Transformation Intensity	Credit Allocation	Mobilization of Liquidity Surplus
Weight ($W_i$)	10.00%	15.00%	60.00%	15.00%

The MR coefficient for each bank can differ and may fluctuate between 8.00% and 17.00% depending on the global score obtained. After calculation, intermediate coefficients are rounded up and vary in steps of 1.0 percentage point.

---

^1 i varies from 1 to 4

---

The coefficient applied to a bank is thus determined based on its relative position on the global score scale. The higher the global score, the closer the coefficient approaches the exemplary bound, indicating better performance. Conversely, a low score leads to a higher, non-exemplary coefficient.

Table 3: Interpretations of Global Scores

Index	Interpretations	MR Coefficient
High Global Score [close to 9]	Bank fully integrated into the new system, effectively contributing to monetary policy transmission and investment financing. → benefits from a lower MR coefficient	8.00%
Low Global Score [close to 0]	Bank showing limited contribution to the system's objectives. → higher MR coefficient	17.00%

---