This paper investigates financial market efficiency in Pakistan, which has been undergoing financial sector reform since the late 1980s in the context of its adjustment and reform efforts. The picture that emerges from the behavior of recently liberalized interest rates in Pakistan raises questions at first about the efficiency of financial markets. A seemingly perverse relationship persists between instruments that, in principle, should command identical returns and the positive valuation of instruments that deliver negative real and effective returns. Rupee-denominated assets, such as administratively controlled bank deposits, federal investment bonds, and treasury bills were paying negative average real returns during the period under consideration. Dollar-denominated foreign currency accounts generally also paid negative exchange rate-adjusted returns.
The paper explores this issue using information contained in the secondary market prices of Foreign Exchange Bearer Certificates (FEBCs), which are rupee-denominated government bonds that must be purchased in foreign exchange but may be cashed in either rupees or foreign currency. FEBCs have been paying ex post rates of return that are, on average, negative in real terms and when adjusted for exchange rate devaluation, and lower than the returns on comparable local and international assets. This perverse interest rate structure challenges efficient market propositions, which constitute the basis of finance theory. This perversity is shown to stem from financial regulation and therefore to be consistent with portfolio optimization and efficient arbitrage of return differentials by market agents.
A reduced-form model for FEBC yield is specified as an equilibrium relation between this yield and the returns on local and foreign substitutes, inflation and devaluation expectations, and regulation proxies. The role of the FEBC as a means of relaxing the legal foreign exchange constraint and of transferring funds from the informal to the formal sector is shown to be a significant predictor of yield. Adjusting for these effects, yield sensitivity to local and international rates of return is sustained. Financial reform appears to have increased the impact on FEBC yield of interest rate movements and reduced those of regulation effects.
The model outperforms conventional interest parity specifications as a description of FEBC yield and, in fact, helps explain why the parity theorem fails to hold for Pakistan. Coefficient estimates of the proxies are used to remove the regulation-induced discount in FEBC yield. Adjusted nominal and real yield series are constructed, indicating that a rational interest rate structure underlies regulation-laden rates of return. The FEBC has been paying an adjusted yield that is positive in real and effective terms and that, on average, is well above LIBOR.