QE Day (2)

  • Stephanie Flanders
  • 5 Mar 09, 01:27 PM GMT

The Bank of England has pressed the button on quantitative easing (see earlier post). The initial size of the scheme - £75bn over the next few months - is smaller than some analysts have suggested, but it's a lot more than if they were dipping their toes in the water.

To give some perspective, £75bn is the equivalent of about 5% of GDP. And the fact that the chancellor has authorised purchases up to £150bn shows that the Monetary Policy Committee (MPC) also knows it may have to do more.

In light of the discussion in my previous post, one quick conclusion I might draw from both the headline amounts and the content of the statements is that the MPC thinks that the asset price (or yield) effect of the policy is going to be more important than the traditional money channel.

I don't know for sure, of course - but, as we saw, if you thought that the money multiplier was in good shape, you would be talking much smaller amounts.

The Bank might demur. Officials tend to say that they don't care how it works if it achieves the right result. By making the purchases over several months, the MPC is also giving itself room to stop, if the effect is much greater than anyone expects.

12/11/08 Bank of England Governor Mervyn King during a press conference at the Bank of EnglandBut in his letter to the chancellor, the Governor, Mervyn King also highlights that £50bn of the total allowed purchases of £150bn should be of corporate securities, "in recognition of the importance of supporting the flow of corporate credit".

Given the size of the two markets and the dangers of taking too much corporate risk onto the Bank's balance sheet, the majority of the purchases will have to be gilts. They don't want to increase the risk premium on government assets, which would push up borrowing rates for everyone and defeat the purpose of the exercise.

But the governor's emphasis is striking. Of course, by emphasising the credit aspects to the policy, King also wants to draw as sharp a contrast as possible between this policy and a policy of printing money simply to finance government deficits. As he will continue to remind us, buying government bonds as part of QE is a means to an end. It is not, as in Zimbabwe, an end in itself.

As I've discussed before, it's an important political and practical difference, even if the short-term effect of what the Bank is doing is the same: namely, monetising part of the government's debt.

QE Day

  • Stephanie Flanders
  • 5 Mar 09, 09:28 AM GMT

This could be the day you've all been waiting for. The day that "quantitative easing" (QE) stops being a rather ugly talking point and becomes a reality.

Bank of EnglandLater today, all the signs are that the MPC will release a statement saying that they have authorised the Bank of England to start buying government and corporate securities on its behalf - paid for with money it has created for itself. The policymakers will probably vote to cut the Bank Rate to half a percent as well. Either way, it will be the move to QE that will be the big news.

I've explained in some previous posts exactly what this policy entails. In effect, by buying the assets, the Bank is trying to engineer an unexpected cash windfall for the institutions selling the bonds to the Bank. If that cash is spent or lent on to other parts of the economy, boosting 'broad money growth' (also known as credit growth) and the amount of activity in the economy, then the purchases will have done their job.

Today's big questions are (a) will it work? And (b) how much will the Bank need to spend? If you polled the brainiest economists in the land, I bet the answers you'd get would be: "maybe" and "we don't know". The demand for money is a fickle thing - it doesn't always do what you want it to do. Monetarists learned that one the hard way in the 1980s.

In his most recent press conference, the Bank's Governor, Mervyn King, seemed pretty confident that QE could work. But even he would admit he has no idea how long it will take - or how much money he will have to print to get there.

Depending on your point of view, there are two big reasons to think that QE will boost the economy. Trouble is, none is guaranteed to work - especially in today's unusual times.

First, and most talked about, is the direct "bank lending" channel for QE. The extra cash in banks' accounts encourages them to lend some of that money out, when that loan gets spent, another bank has more money in its account, some of which they lend out, and so on, and so on - until you end up with a much bigger increase in the amount of broad money (M4) in the economy than the initial deposit by the Bank. That's the "textbook" money multiplier.

Looking at the ratio of narrow money to GDP you might think that multiplier was pretty high - in the region of 15 (so every £1 spent by the Bank would increase the amount of broad money running around the economy by £15). If so, the Bank might not have to spend much at all.

But, we know that banks aren't too keen to lend at the moment. Will having some excess cash on deposit with the Bank of England make them any keener? Maybe not. In that case, the "money multiplier" could be quite small and the Bank would have to spend a lot more.

There's another complication - which is that even if you raised the broad money supply by the desired amount, we don't know exactly how much that will boost national income (GDP). That depends on not just the amount of money around but the speed it's moving around the economy - the velocity of circulation.

In these times of great economic uncertainty and fear about the health of the financial system, the velocity of money will almost certainly be lower than usual, meaning the Bank will need to engineer an even larger rise in broad money to have a given effect on national income, nominal GDP.

That leaves the second channel - the effect on asset prices. There are several different sides to this, but the basic idea is that by purchasing these bonds, the Bank of England will raise the price.

Why? Well, by buying the assets they've reduced the relative supply in the market. They might also have raised liquidity in that market and so made the asset more valuable to investors (this would apply to some corporate bond markets). Or they might have raised the price by increasing demand for those assets indirectly, by giving the sellers some spare cash to spend on something else. (Assuming they do anything with it - they, too, might simply hold on to the cash).

Now, the way bonds work, a higher price means lower borrowing costs for whoever is issuing that debt. Put it another way, if demand has gone up for corporate bonds (for any or all of the reasons above) then companies will need to pay less to raise a given amount of debt.

This asset price effect could be quite powerful, but only if the purchases are large enough to have a significant effect on the price of the assets in question. In the case of gilts, that might be a large number, but it will probably happen.

Of course, simply lowering gilt yields doesn't change much - you've just made it cheaper for the government to borrow, which is already doing plenty of. But you hope that by lowering gilt yields you might make other, higher yielding (corporate) assets more attractive, thereby raising their price as well.

How do you translate all of this theorising into hard numbers? With difficulty. But assume the goal is to raise nominal GDP by around £150bn - that's a common estimate for the shortfall in demand in the economy this year. If you think the money multiplier is alive and kicking and all the banks need is a gentle nudge to lend more, you might think that £10bn in bond purchases would be enough to achieve that.

But, if you think the money multiplier is all messed up because of the credit crunch, and a low velocity of money is going to blunt the policy even more, the ratio will move closer to 1 to 1. And the Bank might have to spend upwards of £100bn to get the desired effect.

It's even harder to produce sensible estimates of the asset price effect. The total stock of outstanding gilts was around £700bn at the end of 2008, and estimates of the amount to be issues this year are rising almost by the week. But how much the Bank affects gilt prices (and yields) through its purchases will be down to market psychology as much as straight arithmetic.

That poses a quandary of its own: the Bank doesn't want to overdo it on its first shopping spree, but if it seems too tentative about the whole endeavour, investors are less likely to think there's been a lasting change in the demand for bonds, and prices won't change.

Where does the Bank come out in all this? The phrase I have heard used by Bank officials more than once is "suck it and see". Given the nature of the stakes, you might find that a bit disturbing. I call it an honest assessment of where we are.

Update 1327: More, and a response to your comments in this post.

The BBC is not responsible for the content of external internet sites