Like a dipsomaniac who drinks every day, markets might have forgotten what it’s like not to wake up with a hangover and reach for a Bloody Mary. For almost a decade, we’ve been living with the hangover from the mid-2000s asset bubble — the bender to end all benders. But for years now, markets and economies have had ready access to a hair-of-the-dog: super-accommodative monetary policy.

You can point to the post-financial crisis bull market — in stocks and in bonds — and say it’s based on fundamentals, I guess. But there can be little doubt that it wouldn’t have climbed so far or so fast without the support of the twin pillars of monetary stimulus: ultra-low interest rates, and huge exercises in bond-buying from four of the biggest central banks in the world — the Bank of Japan, the European Central Bank, the Bank of England and the U.S. Federal Reserve.

So what happens to the drunk when you take away his hangover cure? Or, to leave aside the boozy metaphor, what happens when central banks withdraw stimulus?

Central banks

We’re about to find out, because the monetary pillars are showing cracks. The Bank of Japan alone seems committed to its low rate/quantitative easing program, but other central banks are not so firm. The Bank of Canada has signalled that it’s in the mood to remove the 50 basis points of stimulus it injected in 2015; the Bank of England has mused that it’s time to raise rates; ECB head Mario Draghi has started talking about “adjustments” in monetary policy. The Fed — which has already raised rates twice this year — is further ahead than any of them, and it’s about to do what the other big boys have not: unwind the legacy of quantitative easing.

While the Fed has been on a tightening path for some time, the unwinding of its balance sheet is a relatively new wrinkle, but it could be a biggie. Back in 2007, the Fed had only about US$870 million on its balance sheet. As of June 28 of this year, it was carrying US$4,516,885,000,000. (That’s about US$4.5 trillion in easier-to-type terms.) What’s in there? Some is just regular old central bank stuff — gold stock (US$11 billion), foreign currency-denominated assets (US$21 billion), and so on. But much of the rest is made up of what you might call Markets’ Little Helpers.

The biggest dose is U.S. Treasuries, which the Fed bought between 2008 and October 2014 in three waves of quantitative easing, effectively increasing the money supply. The Fed’s Treasury trove sits at about US$2.5 trillion; by comparison, the total outstanding Treasury market is valued at about US$14 trillion. The other chunk comprises mortgage-backed securities, which the Fed bought up to encourage lenders to issue mortgages. MBS comprise about US$1.8 trillion on the Fed’s balance sheet — that’s about 20 per cent of total outstanding mortgage-related debt securities in the U.S. market.

Now, chair Janet Yellen and her counterparts on the Federal Open Market Committee have decided to at least get ready to unwind that big chunk of debt. It will be a delicate balancing act. Move too fast or too far, and the hangover that quantitative easing was supposed to cure could come back — with a vengeance — in the form of soaring bond yields/falling bond prices, higher borrowing costs and economic stagnation.

The Fed

We still don’t know when the Great Unwinding is going to start; the minutes of the FOMC’s June meeting, released on Wednesday, provided no clues, other than it will be sometime this year and markets might be given “a couple months” advance notice. But the Fed has been pretty clear on the how.

Clearly, the Fed’s goal is to minimize market disruption, so it’s going to unwind gradually and passively. Actively selling more than US$4 trillion in debt securities on the open market could easily spark a run on Treasuries and lead to soaring mortgage rates for American consumers. Instead, the Fed is simply going to stop buying enough new bonds to fully replace current holdings as they mature. That will increase supply only indirectly, by withdrawing demand, which the Fed hopes will limit the impact.

The Fed is also going to limit the balance sheet reduction to pre-set monthly caps, which would gradually rise after a year to US$30 billion a month for Treasuries and to US$20 billion a month for MBS. Back-of-the-envelope, that means it would take about seven years to reduce Treasury holdings to zero, and about eight years for MBS.

But the Fed is not going that far. It plans to unwind only part of the total balance sheet. In the end, and if everything goes according to plan, the unwinding will result in an “appreciably” smaller balance sheet, but one still bigger than it was before the recession.

That is how it is supposed to work, anyway. Yet as predictable as the Fed wants the unwinding to be, there will no doubt be side effects. Gradually or no, one humongous buyer is withdrawing from the bond market. At the very least, there will be, as central bankers like to say, a period of “adjustment” — which investors might feel in the form of higher rates and higher volatility, and which American consumers and businesses might feel in rising borrowing costs.

We just don’t know how difficult that adjustment will be. If post-recession stimulus was a great, and not altogether successful, experiment in central banking, withdrawing that stimulus will be just as big an experiment — and one just as fraught with risk.