TY - JOUR AU - Obstfeld,Maurice AU - Rogoff,Kenneth TI - Ruling Out Nonstationary Speculative Bubbles JF - National Bureau of Economic Research Working Paper Series VL - No. 1601 PY - 1987 Y2 - April 1987 UR - http://www.nber.org/papers/w1601 L1 - http://www.nber.org/papers/w1601.pdf N1 - Author contact info: Maurice Obstfeld Department of Economics University of California, Berkeley 530 Evans Hall #3880 Berkeley, CA 94720-3880 Tel: 510/643-9646 Fax: 510/642-6615 E-Mail: obstfeld@econ.berkeley.edu Kenneth S. Rogoff Thomas D Cabot Professor of Public Policy Economics Department Harvard University Littauer Center 216 Cambridge, MA 02138-3001 Tel: 617-495-4022 Fax: 617/495-7730 E-Mail: krogoff@harvard.edu AB - There is a large and growing empirical literature that tests forthe existence of asset-price bubbles or "sunspot" equilibria -- equilibria unrelated to market fundamentals. Our view is that even tests for non-stationary asset-price bubbles should not be interpreted as such. In the present paper we extend earlier work of ours which provided a strong case for ruling out nonstationary speculative price bubbles in models based on individual maximizing behavior. In the first part of the paper we study the possibility of stochastic exploding price-level bubbles of the kind proposed by Blanchard (1979). As in our previous work, a scheme of fractionally backing the currency with real outputis sufficient to preclude such bubbles. In the second part of the paper we examine conditions for ruling out implosive price-level bubbles, equilibrium paths along which the price level asymptotes to zero even though the monetary growth rate is constant. A condition on preference simplied by any reasonable monetary transactions technology is sufficient to prevent such bubbles from emerging. Given that anticipated future disturbances can lead to price paths which are qualitatively indistinguishable from bubble paths, and given the strong theoretical basis for ruling out nonstatioflary bubbles, our conclusion is that any "positive" evidence of bubbles should be regarded only as evidence of omitted variables. ER -