The Efficient Market Hypothesis in Personal Finance: Choosing an Adjustable or a Fixed Rate Mortgage

Article excerpt

ABSTRACT

Homebuyers and commercial real estate investors who obtain funds using mortgages face the choice of a fixed or an adjustable rate mortgage. Fixed rate mortgages have constant payments, but a high initial rate: adjustable rate, interest-only or hybrid mortgages begin with lower rates, but change at fixed intervals over time. Using a straight forward forecasting model, we find that short rates can be forecasted. However, given the limited duration of the cost advantage in the study period, and the transaction and penalty costs of adjusting a borrowing strategy, the decision may not be economically significant.

Introduction

Mortgage borrowers who face the decision of a fixed or adjustable rate mortgage are often given the rule of thumb that if their horizon is shorter than five years they should consider an adjustable rate mortgage. Lengthy horizons traditionally mandate a fixed rate, long-term mortgage. The rationale is that adjustable rate mortgages (ARM) are cheaper in the short run, but have the drawback of potentially rising rates. Therefore borrowers who plan to move or refinance in a relatively short time frame will never face the large rate increases. Some authors have even claimed that all borrowers, regardless of horizons, should choose adjustable rate mortgages because the early-expected savings are worth the later risk [Longhofer (2006)].

Explicitly or implicitly, borrowers must estimate the present value of the expected borrowing costs of each type of mortgage and choose the lowest cost. The actual present value will depend on the size of the loan, the transaction cost at the beginning, both financial and transfer costs, and the actual length of the holding period. Since ARMs have low early rates, short horizon periods are biased in their favor. However, even with short periods, ARM rates can rise. With a fixed rate mortgage, the interest rate and payment is set for the life of the mortgage.

In the received finance literature, Modern Portfolio Theory, beginning with Harry Markowitz (1959) led to what is now referred to as the Efficient Market Hypothesis (EMH). The theory posits that investors cannot expect abnormal investment returns over the long run, because as information emerges, it is impounded in stock prices. The investor does not even have time to act on the new information before stock prices change. As applied to the present topic, choices between ARMs and fixed rate mortgages, EMH would imply that mortgagors could not gain an advantage by choosing one type over another. The present value of both types of mortgage costs would be equivalent, even if mortgagors could forecast interest rates, because market expectations of future interest rates would be immediately discounted into current interest rates.

This paper develops a forecasting model for rates and compares the forecast with actual rates from 2006 to 2007. We apply a forecasting model based on past rates to ascertain if future movements can be predicted in an economically significant way. The comparison shows that if the forecast model had been used by retail-level borrowers, little behavior change would result due to transaction costs and the implementation Costs of modeling Information on expected future interest rates has been discounted into current interest rates.

Literature Review

Until 1982, restrictions on federally insured depository institutions limited offerings to long term, fixed rate equal payment mortgages. Subsequently, innovations known as adjustable rate mortgages (ARMs) were introduced as volatile and rising interest rates threatened the capital integrity of financial intermediaries. The goal of the new mortgages was originally to protect lenders from interest rate risk in times of rising rates. Since the risk of rising rates is, at least partially, transferred to borrowers, ARMs have introductory rates and payments which are less than those of fixed rate mortgages for comparable borrowers at the time of origination. …