Approaches to modeling liquidity maturity

Savings and liquidity

Approaches to modeling liquidity maturity

Savers who place their cash in deposit accounts – and can withdraw the funds from their account at any time – are thus able to decide the liquidity profile of this important source of funding for banks. This is a very important consideration for bank risk managers. In an article about savings in the previous edition of Zanders Magazine, we showed that banks must include adequate modeling of the liquidity maturity as part of their integral liquidity management. But how should this modeling be approached?

Savings are a stable source of funding for banks. On the one hand, many savers view savings as a simple and accessible investment instrument while, on the other hand, it is a safe form of investment due to the deposit guarantee scheme. However, risk managers have to take account of a special characteristic: the saver may at any given moment withdraw this funding. In other words: it’s the savers themselves who decide the liquidity profile of the bank’s funding. As adequate modeling of liquidity maturity is an essential part of any bank’s integral liquidity management, we have summarized the options.

Outflow model versus volume model

The liquidity maturity is determined on the basis of a liquidity cash flow chart; a chart which is determined by outflow models and volume models. Assuming a ‘dead portfolio’, outflow models estimate the liquidity profile of the savings of their current clients. By only modeling on the basis of outflow and ignoring inflow, the assessment of volume development is prudent. The periodic outflow is estimated with the help of a so-called vintage analysis. This technique randomly selects a point in the savings portfolio at a number of different moments in time. Each group of clients selected form a vintage, for which a descending series of (volumes of) savings are determined by measuring the periodic outflow (see Figure 1).

Savings fig1

Figure 1: Historical series of (volumes of) savings and the outflow of different vintages.

Frequently, it is evident that the outflow percentages decline over the course of time. This time effect is caused by a relatively high outflow from unstable clients with volatile savings balances, after which, a stable core remains. In addition, certain patterns are sometimes observed in the series of outflows and these can be included as explanatory variables. Examples of these include seasonal effects, the impact of competitors’ savings rates and/or macro-economic variables.

Under the ‘going concern’ principle, volume models estimate the development of the entire savings portfolio. Implicit in this model is that a role is played by the outflow and inflow of savings from both existing and new clients. The volume model simulates large numbers of volume paths, on the basis of which a worst case (volume of) savings series is determined (see Figure 2). In this context, the confidence level indicates the degree of prudence in the risk modeling.

savings fig 2

Figure 2: Simulated volume paths and worst-case (volume of) savings series

Model segmentation

In the modeling of liquidity cash flow profile, the segmentation of the savings portfolio is important. The degree of segmentation depends on the materiality and heterogeneity of the savings. This is because the attitude to liquidity varies enormously between different product types and client groups. Even within client groups, certain factors may require further segmentation. For example, the level of savings, due to the coverage limit of the deposit guarantee scheme.

Choice of model

It is important that the model chosen is in line with the way in which other balance sheet items are accounted for in liquidity management. For example, combining the assumption of a dead portfolio for the liquidity cash flow profile of mortgages with a goingconcern principle for savings will lead to inconsistencies. Such inconsistencies could cause an incorrect assessment to be made of a bank’s liquidity requirements. In practice, however, the availability of data largely determines the choice of model.

Outflow models require significantly more data. It is self-evident that the advantage of a detailed dataset is that it can chart the liquidity profile more adequately; which in turn means the bank is better placed to forecast and control its liquidity position.

More articles about ‘The modeling of savings’: