One sure-fire way to increase profitability is to deploy excess liquidity into loans. On a risk-adjusted basis, loans are superior to securities in producing returns. All bankers know this. Institutions with high loan-to-asset ratios tend to have better profitability than institutions with smaller loan portfolios as shown at right. Banks in Indiana and Michigan with ROEs greater than 16% have consistently held far more loans as a percentage of assets than lower ROE institutions.

So why don’t all institutions maintain high loan-to-asset ratios to maximize returns? In most cases the answer is self-imposed policy limits. Policy limits should be set based on a cost/benefit analysis. Restrictive policies based on loans to deposits or mandating the sale of fixed-rate loans may result in a lower risk profile, but not without cost. Policy limits that reduce loan-to-asset ratios will also reduce profitability.

Barbell funding matches risk

Barbell funding strategies are extremely flexible. The funding structure can be customized to match the interest rate risk objective. For example, an institution may be averse to adding fixed-rate mortgages longer than five-year balloons. Barbell funding can be used with longer term mortgages to create an interest rate risk profile similar to five-year balloons. This is accomplished by funding the first five years of the mortgages’ maturities with existing liquidity and the remaining maturities with long-term FHLBI borrowings. Consider the case of current 15 year mortgage production.

During early March 2000, the market rate for 15 year mortgages was approximately 8.25%. While infinite barbell funding strategies are possible, a structure of 70% liquidity (to fund the first five years) and 30% six- and seven-year FHLBI borrowings could be used. Assuming the liquidity was previously invested in fed funds or short-term securities at 5.75% and the FHLBI borrowing cost was 7.30%, the weighted average funding cost would be 6.22% (the liquidity funding cost is an opportunity cost). In other words, the strategy would produce a 203 basis point spread at the start (8.25% - 6.22%). While the strategy looks good initially, how would it affect the institution’s risk?

If interest rates were to rise, long-term funding would provide a significant hedge. While only a fraction of the funding is fixed rate (30%), there is enough to anchor the overall cost of funds through most rate fluctuations. The cost of funds can only change 70% of any rate change, since 30% of the cost is fixed. This allows the barbell strategy to sustain nearly a 3% rate shock and still maintain a positive spread. Increasing the percentage of long-term funding provides even greater protection against rising interest rates. Of course, the cost of this protection is a higher initial funding cost and a lower initial spread. Also, increasing the percentage of fixed rate funding reduces the protection against the risk of falling interest rates proportionately.

If interest rates were to fall, the liquidity portion of the funding strategy would result in a wider spread. An immediate 2% drop in rates would hypothetically widen the spread to 343 basis points. Unfortunately, this spread would be short-lived since prepayments would accelerate. However, managing prepayments is a strength of barbell strategies. Incorporating only 30% fixed funding helps prevent prepayments from reducing the loan balance below the funding level.

Effect on other risks modest

The barbell funding strategy also has an impact on both credit and liquidity risks. The strategy results in a modest increase in both credit risk and leverage. In effect, the strategy substitutes 50% risk-weight mortgages for 20% risk-weight fed funds or securities. Risk assets will be increased by the strategy. However, since conforming Midwestern mortgages historically have exhibited very high credit quality, the actual increase in credit risk is probably slight.

Finally, the strategy would have a modest impact on liquidity risk. While fed funds would be reduced and FHLBI borrowing capacity would be used, the new mortgages could be pledged for future borrowings. The table below shows the net impact on overall liquidity.

This article has been presented for educational purposes only. The FHLBI is not a financial or investment advisor. It is solely the reader's responsibility to evaluate the risk and merits of any funding strategy or business proposal.