About 50% of all lenders in the Dutch mortgage market that are offering new mortgage loans provide automatic risk class adjustment. This means that, as the LTV of the loan decreases due to (p)repayments, the interest rate that is applicable to the loan also decreases. Inevitably, this impacts the returns of Dutch mortgage loan investors. But by how much? And what does this mean for the valuation of the portfolio? These are the questions that we will answer in this article. And as it turns out, this also affects the valuation of portfolios with loans that do not have automatic risk class adjustment.

A mortgage loan is said to have Automatic risk class adjustment (ARCA) if the lender automatically adjusts the interest rate to a lower rate when the loan reaches a lower risk class (LTV segment). This product feature has become relatively standard in the Dutch mortgage market over the past years. It favors the borrower, but what does this mean for the investor’s return? And how does it impact the portfolio valuation? These questions will be answered in this article.

Currently, about 50% of the available products in the market offer ARCA. Lenders that don’t offer this product feature yet have to be more competitive in their pricing to win market share as mortgage advisors often consider the total cost of the mortgage (including future rate drops) rather than just looking at the initial interest rate.

To understand the implications of this product feature, it is essential to consider what variables affect the impact of ARCA, which are: 1) the number of risk classes and their boundaries, 2) the current LTV of the loan, 3) the rate difference between risk classes, and 4) the speed at which a loan repays. In the following sections we illustrate how each of these four components impacts the yield of a loan.

1. The number of risk classes and their boundaries

As the number of risk classes increases, the number of interest rate drops increases which naturally leads to a lower yield. In addition to that, it is also the timing of the rate drop that has an impact on the yield. The timing of the rate drop is affected by the boundaries of the LTV segments (in addition to other factors which are covered later). To illustrate these effects, consider an annuity loan with a 30-year fixed rate period. We assess the yield of this product for three different sets of risk classes:

• Situation 1: LTV > 90%, 80% < LTV <= 90%, 70% < LTV <= 80%, 60% < LTV <= 70% and LTV <= 60%;
  

  • Situation 2: LTV > 90%, 60% < LTV <= 90% and LTV <= 60%, and;
    

    • Situation 3: LTV > 90%, 70% < LTV <= 90% and LTV <= 70%.
      

The interest rate of this loan for the three situations over time is illustrated in Figure 1 below.

Figure 1: Interest rate scenarios for an annuity loan with a 30-year fixed rate period and different LTV buckets. Source: LoanClear

As expected, Situation 1, which has the most rate drops, has the lowest product yield (1.628%). Situation 2 has the highest yield (1.711%) as it has less rate drops than Situation 1 and a lower boundary for the lowest risk class resulting in the last rate drop to occur 38 months later than for Situation 3 (1.637%).

2. The current LTV of the loan

The current LTV of the loan is very important for determining the yield impact as it also affects the timing of the rate drops and how many rate drops are left to take place. To illustrate these effects, again consider an annuity loan with a 30-year fixed-rate period. The product has five risk classes namely: LTV > 90%, 80% < LTV <= 90%, 70% < LTV <= 80%, 60% < LTV <= 70% and LTV <= 60%. Now consider the following three situations:

• Situation 1: The starting LTV of the loan is 100%;
  

  • Situation 2: The starting LTV of the loan is 91%, and;
    

    • Situation 3: The starting LTV of the loan is 75%.
      

The interest rate of this loan for the three situations over time is illustrated in Figure 2 below.

Figure 2: Interest rate scenarios for an annuity loan with a 30-year fixed rate period and different starting LTVs. Source: LoanClear

Both Situation 1 and Situation 2 start in the same LTV bucket and, therefore with the same interest rate. However, the yield for Situation 1 (1.628%) is higher than that of Situation 2 (1.512%) because the interest rate drops in Situation 2 take place earlier due the lower starting LTV. On the other hand, Situation 3 has a much lower starting LTV and because of that starts in a lower risk class. This results in a lower starting interest rate and consequently in a lower yield (1.330%). The final yield differs less from the starting rate than for Situation 1 and 2 as less rate drops take place.

3. Rate difference between risk classes

Perhaps more obvious is the impact of rate differences between different risk classes. Consider again the same annuity loan with a 30-year fixed rate period. The product has five risk classes namely: LTV > 90%, 80% < LTV <= 90%, 70% < LTV <= 80%, 60% < LTV <= 70% and LTV <= 60%. Now consider the following two situations:

• Situation 1: The rates for the different risk classes are 2.00%, 1.80%, 1.60%, 1.40% and 1.20% respectively, and;
  

  • Situation 2: The rates for the different risk classes are 2.00%, 1.70%, 1.40%, 1.10% and 0.80% respectively.
    

The interest rate of this loan for the two situations over time is illustrated in Figure 3 below.

Figure 3: Interest rate scenarios for an annuity loan with a 30-year fixed rate period and different rate drop sizes. Source: LoanClear

The yield for Situation 1 (1.628%) is higher than for Situation 2 (1.427%). Traditionally, a loan tape only shows the rate that is currently applicable to the loan. Therefore, the loans in Situation 1 and 2 might seem identical at first but, after more careful analysis, turn out to be quite different in terms of yield.

Another interesting effect here, even though minor, is that because of the lower interest rates for Situation 2, the rate drops for Situation 2 take place slightly earlier because more principal is being repaid. So In this specific example, the time difference is 2 months at most.

4. Speed at which the loan repays

As discussed in section 1, the sooner a loan reaches the next risk class, the lower the yield. This is partly determined by the boundaries of the risk classes but also by the speed at which the loan is being repaid. Repayment speed is determined by the payment type, the interest rate, the loan term and the prepayment rate. Consider a 30-year fixed-rate period loan with the following risk classes: LTV > 90%, 80% < LTV <= 90%, 70% < LTV <= 80%, 60% < LTV <= 70% and LTV <= 60% and a starting LTV of 100%. Now consider the following four situations:

• 100% annuity loan, without assuming prepayments;
  

  • 50% annuity loan and 50% interest-only loan without assuming prepayments, and;
    

    • 100% annuity loan, assuming a 3% prepayment rate;
      

      • 50% annuity loan and 50% interest-only loan, assuming a 3% prepayment rate.
        

The interest rate for these situations over time is illustrated in Figure 4 below.

Figure 4: Interest rate scenarios for a loan with a 30-year fixed rate period and repayment speeds. Source: LoanClear

Yields for Situation 1 to 4 are 1.628%, 1.749%, 1.518% and 1.529% respectively. It can clearly be seen that lower repayment speeds result in a later rate drop and therefore a higher yield.

How does automatic risk class adjustment impact valuations?

From the four illustrations above, it is clear that many factors impact the yield for a product with automatic risk class adjustment. The effects are also quite significant, with the difference between a 100% annuity loan and a 50% annuity and 50% IO being 12 bps.

When looking at these loans from a valuation perspective, we see that they are negatively impacted as a lower product yield results in lower interest cash flows and therefore a lower valuation assuming all else equal.

This is not the only effect however, as discount rates are also impacted. Many valuation methodologies use primary market rates for determining the discount rates with which the loans are valued. In the analyses above we found that the product yields for loans with automatic risk class adjustment are lower than the initial rate that is oberserved. When this effect is taken into account, it will cause primary market rate estimates to decrease also. This will lead to lower discount rates and then again to higher valuations.

The exact valuation impact for a specific loan will be dependent on the four factors discussed in this article but also on these four factors of all the other products in the market that operate in this same space as this will influence how the discount rate is affected. On average though, we see lower valuations for loans with automatic risk class adjustments and slightly higher valuations for loans without automatic risk class adjustment compared to a valuation methodology where automatic risk class adjustment is ignored.

Conclusion

Automatic risk class adjustment is more and more becoming the standard in the Dutch mortgage market and is impacting returns and valuations for investors. Many factors contribute to this and it is important for investors to understand how these dynamics work and how they affect them. Specifically for the valuation of these assets it is an important item as it does not only affect investors that invest in loans with automatic risk class adjustment but all investors that make use of primary market rates for the valuation of their investment.

Disclaimer

Please read our full disclaimer here.