3 i4 THE GENERAL THEORY OF EMPLOYMENT bk. vi

propelling it upwards at first gather force and have acumulative effect on one another but gradually losetheir strength until at a certain point they tend to bereplaced by forces operating in the opposite direction;which in turn gather force for a time and accentuateone another, until they too, having reached their maxi-mum development, wane and give place to their op-posite. We do not, however, merely mean by a cyclicalmovement that upward and downward tendencies, oncestarted, do not persist for ever in the same directionbut are ultimately reversed. We mean also that thereis some recognisable degree of regularity in the time-sequence and duration of the upward and downwardmovements.

There is, however, another characteristic of what wecall the Trade Cycle which our explanation must coverif it is to be adequate; namely, the phenomenon of thecrisis —the fact that the substitution of a downwardfor an upward tendency often takes place suddenly andviolently, whereas there is, as a rule, no such sharpturning-point when an upward is substituted for adownward tendency.

Any fluctuation in investment not offset by a corre-sponding change in the propensity to consume will, ofcourse, result in a fluctuation in employment. Since,therefore, the volume of investment is subject to highlycomplex influences, it is highly improbable that allfluctuations either in investment itself or in the marginalefficiency of capital will be of a cyclical character. Onespecial case, in particular, namely, that which is associ-ated with agricultural fluctuations, will be separatelyconsidered in a later section of this chapter. I suggest,however, that there are certain definite reasons why,in the case of a typical industrial trade cycle in the nine-teenth-century environment, fluctuations in the mar-ginal efficiency of capital should have had cyclical char-acteristics. These reasons are by no means unfamiliareither in themselves or as explanations of the trade