secure basis there has now been piled the most chancy and insecurely experimental system of conventions and assumptions about money and credit it is possible to imagine.  There has grown up a vast system of lending and borrowing, a world-wide extension of joint-stock enterprises that involve at last the most fantastic relationships.  I find myself, for example, owning (partially, at least) a bank in New Zealand, a railway in Cuba, another in Canada, several in Brazil, an electric power plant in the City of Westminster, and so on, and I use these stocks and shares as a sort of interest-bearing money.  If I want money to spend, I sell a railway share much as one might change a hundred-pound banknote; if I have more cash than I need immediately I buy a few shares.  I perceive that the value of these shares oscillates, sometimes rather gravely, and that the value of the alleged money on the cheques I get also oscillates as compared with the things I want to buy; that, indeed, the whole system (which has only existed for a couple of centuries or so, and which keeps on getting higher and giddier) is perpetually swaying and quivering and bending and sagging; but it is only when such a great crisis occurs as that of 1907 that it enters my mind that possibly there is no limit to these oscillations, that possibly the whole vast accidental edifice will presently come smashing down.

Why shouldn’t it?

I defy any economist or financial expert to prove that it cannot.  That it hasn’t done so in the little time for which it has existed is no reply at all.  It is like arguing that a man cannot die because he has never been known to do so.  Previous men have died, previous civilisations have collapsed, if not of acute, then of chronic financial disorders.

The experience of 1907 indicated very clearly how a collapse might occur.  A panic, like an avalanche, is a thing much easier to start than stop.  Previous panics have been arrested by good luck; this last one in America, for example, found Europe strong and prosperous and helpful.  In every panic period there is a huge dislocation of business enterprises, vast multitudes of men are thrown out of employment, there is grave social and political disorder; but in the end, so far, things have an air of having recovered.  But now, suppose the panic wave a little more universal—­and panic waves tend to be more extensive than they used to be.  Suppose that when securities fall all round, and gold appreciates in New York, and frightened people begin to sell investments and hoard gold, the same thing happens in other parts of the world.  Increase the scale of the trouble only two or three times, and would our system recover?  Imagine great masses of men coming out of employment, and angry and savage, in all our great towns; imagine the railways working with reduced staffs on reduced salaries or blocked by strikers; imagine provision dealers stopping consignments to retailers, and retailers hesitating to give credit.  A phase would arrive when the police and militia keeping order in the streets would find themselves on short rations and without their weekly pay.