Every quarter, Latvijas Banka assesses cyclical risks and the adequacy of the CCyB rate and publishes the assessment. If necessary, a decision on changing the CCyB rate is taken. The assessment takes account of the Recommendation of the European Systemic Risk Board (hereinafter – ESRB) of 18 June 2014 on guidance for setting countercyclical buffer rates (ESRB/2014/1)(hereinafter – Recommendation).
According to the Recommendation, the deviation of the credit-to-GDP ratio from the long-term trend in relation to loans granted to the private non-financial sector is used as a starting point for the CCyB rate assessment. An increasing, positive deviation of the credit-to-GDP ratio from its long-term trend suggests that lending has reached excessive levels, posing risks to the financial system. To calculate the long-term trend of the credit-to-GDP ratio, the one-sided Hodrick-Prescott filter[1] with the smoothing parameter value λ = 400 000 is used according to the Recommendation.
It is important to choose the relevant loan indicator when performing an assessment. The loan definition proposed by the Basel Committee on Banking Supervision and used to calculate the standardised credit-to-GDP gap includes the liabilities of the private non-financial sector to both credit institutions and non-bank financial institutions (the "broad" definition of loans). According to the Recommendation, Member States may in addition employ an alternative lending indicator provided it displays better signalling qualities. Latvia uses a more narrowly defined lending indicator as a more appropriate additional indicator. The above lending indicator comprises only loans granted by credit institutions and the purchased non-financial sector debt securities (the "narrow" definition of loans).
The assessment also depends crucially on the selection of the starting point of data time series. In the case of Latvia, it should be borne in mind that the data time series is relatively short and that significant structural changes have affected data. According to estimates, the simulation of setting the historical CCyB rate in Latvia is the closest to experts' assessments regarding the credit cycle if the year 1999 is used as the starting point of the time series.
Using the calculated credit-to-GDP gap, the benchmark buffer rate is determined as fallows - where the credit-to-GDP gap equals or is below 2 percentage points, the benchmark buffer rate is 0%. Where the gap exceeds 2 percentage points, the benchmark buffer rate linearly (according to the Recommendation) increases from 0% to 2.5% (the upper threshold) of risk-weighted assets when the credit-to-GDP gap reaches and exceeds 10 percentage points. Benchmark buffer rate is calculated based on both the standardised credit-to-GDP gap as well as the additional (narrow) credit-to-GDP gap. Taking into account that Latvia uses additional (narrow) credit definition as a more appropriate approach for starting point to decisions underlying CCyB rate, the benchmark buffer rate calculated using narrow credit data is also the buffer guide.
However, the calculated CCyB guide is not the only reference point when assessing the adequacy of the CCyB rate. Comprehensive additional quantitative and qualitative information helping to assess the cyclical systemic risk is also taken into account. Pursuant to the Recommendation, the assessment takes account of additional indicators characterising lending development, potential real estate repricing, external imbalances of the economy, strength of credit institutions' balance sheets, private debt burden, as well as potential mispricing of risks. These indicators, in addition to the CCyB assessment, are published on Latvijas Banka's website on a quarterly basis. Latvijas Banka has also developed a composite cyclical risk indicator used to assess the cyclical risk.
The CCyB rate is expressed as a percentage of the total value of exposures to Latvian residents between 0 and 2.5% (in justified cases, a higher CCyB rate, exceeding 2.5%, may be set).
In the event of raising the CCyB rate, credit institutions have to start maintaining it within 12 months following the publication of the respective notification, but a shorter application period is possible in certain cases. The reduction in the CCyB rate enters into force once the respective decision is taken.
Credit institutions also have to take account of the CCyB rates set in other countries provided they have exposures in the respective countries. Therefore, each credit institution has to calculate its specific CCyB rate. It is calculated as the weighted average rate taking into account the geographical breakdown of exposures and the CCyB rate set in the respective countries. The obtained weighted average credit institution-specific CCyB rate is multiplied by the total value of exposures. The CCyB requirement has to be met by Common Equity Tier 1 capital.
[1] The Hodrick-Prescott filter is a mathematical tool used to establish the long-term trend of a variable. Parameter λ = 400 000 corresponds to the financial cycle that is longer than the economic cycle.