Posts tagged ‘monetary policy’

20 May 2012

How the Bank stole the recovery! (part 1)

I flagged earlier that I wanted to explore some reasons why quantitative easing (QE) hasn’t succeeded in prompting a broader recovery in the UK.

To start off with, let’s think about how QE is supposed to work. There are a number of possible transmission channels – for instance, I took the following diagram from the Bank of England’s Q3 2011 Quarterly Bulletin:

However, I want to focus on a transmission channel that the Bank (given its 2%-inflation mandate) has been relatively quiet about – the impact of QE through nominal variables. I’m happy to be agnostic for now about exactly how this channel is supposed to work, but there are two broad sets of ideas:

  • Keynesians / New Keynesians (such as Paul Krugman) often emphasise the importance of real interest rates in a deflationary environment. Given that nominal interest rates cannot go below zero, a sufficiently-positive rate of inflation is required to achieve the negative real interest rate that a depressed economy might need in order to recover.
  • Market monetarists (such as Scott Sumner) focus on nominal spending as the determinant of aggregate demand. In this view, a central bank can always increase nominal spending by debasing the currency; and, at times when the economy isn’t supply-constrained, growth in nominal GDP (NGDP) is a precondition for growth in real GDP.

Whichever of these views you might hold though, it’s clear that the ability of the central bank to act in an inflationary manner is crucial. In the New Keynesian view, you need to increase inflation expectations so as to reduce real interest rates (since r = i – πe). In the market monetarist view, you need to increase expectations of nominal spending, of which inflation expectations are a component. Either way, to use Krugman’s phrase, the central bank needs to ‘credibly promise to be irresponsible’.

Let’s see how that’s working out for the Bank:

Looking at the 5-year average inflation rates implied by market prices, we can see that QE and a rise in inflation expectations during 2009 did occur together. But also that: (a) inflation expectations started rising from their post-Lehman trough two months before QE started; and (b) more crucially, inflation expectations merely returned to their pre-crisis range of 2-3%, rather than the ‘irresponsible’ levels a central bank might need to target to reduce real interest rates / boost nominal spending, as a depressed economy might require.

Partly, the Bank’s failure to be credibly inflationary is the flip side of the successful introduction of inflation targeting in the UK in the 1990s. People don’t believe that a central bank legally committed to achieving 2% inflation is likely to let prices get substantially ahead of that.

But to some extent, this is also due to the Bank actively telling people that it won’t ‘irresponsibly’ allow QE to be inflationary – even though allowing QE to be inflationary may be crucial to allowing QE to succeed! Here’s Mervyn King at the Treasury Select Committee in February of this year (my bold):

I think actually unwinding [QE] is one of the more straightforward aspects. Either we would sell the gilts that we had purchased back on to the market or if we felt we were putting too many gilts on the market at the same time, or in a short period, we could issue short-term securities, Bank of England bills, and sell those. Actually mopping up the liquidity is a relatively straightforward aspect of this process. I am in no doubt that if we felt the need to unwind the economic effects of asset purchases, it would be a lot easier to slow the economy down when we come to unwind it than it is now to try to encourage growth in the economy by conducting asset purchases. I think going the other way would be a lot easier than trying to stimulate the economy at present.

So what he’s telling people, very clearly, is that the Bank is willing and able to unwind the large increase in the monetary base. In other words, the Bank views QE as a temporary increase in the monetary base – but then why should people assume it will have a permanent effect on the future price level? The Bank is not credibly promising to be inflationary.

This is why I’m increasingly attracted to targeting the level of NGDP, rather than the annual rate of inflation. If we can get the Bank to credibly commit to a future path for the level of NGDP in each of the next few years – with a legal mandate for the MPC to hit that path, irrelevant of what that means for inflation – then I think we’d go a long way to addressing the demand deficiency we’re experiencing. Either nominal spending does drive real spending, in which case hitting the NGDP target leads to real growth too; or nominal spending doesn’t directly drive real spending, in which case hitting the NGDP target means price inflation, which reduces real interest rates and in turn does drive real growth.

But in the meantime, the Bank still has a 2% inflation target – which the market believes it can maintain, in spite of QE, because the Bank is telling everyone that QE will be unwound in the future.

So no wonder all that QE doesn’t seem to have done much for us yet!

Advertisements
20 May 2012

Chart of the week: 14-20 May 2012

I’ve taken this week’s from the Bank of England’s latest Inflation Report.

In spite of unprecedented monetary stimulus, the Bank still expects CPI inflation to actually undershoot the 2% target in the next few years. I’ve previously suggested that this is because the Bank’s approach to quantitative easing has focussed on buying assets that are close substitutes for monetary base – which may have contributed to a lack of feed-through into broad money.

But other factors could also be at play – and I want to explore some of these in the next few days.

14 May 2012

Yes, sovereign currency issuers can always be ‘solvent’ – but don’t be so damned blasé about it

This video on New Economic Perspectives explains the view of modern monetary theorists (MMTers) that a ‘monetary sovereign’ – i.e. one with its own domestically-controlled currency, like the UK or US, rather than a country in the euro area – does not face financial constraints and therefore cannot be forced to default on domestic-currency debts.

I’ve heard this view expressed a number of times to explain why either: (a) the ratings agencies are stupid to downgrade the debts of countries like the US;  (b) governments of sovereign currency issuers are stupid to pursue fiscal consolidation in countries like the UK; or (c) euro members are stupid to have foregone monetary sovereignty by joining the single currency.

The argument behind this is incredibly simple. It goes like this.

(1) Monetary sovereign governments in advanced economies spend, tax and borrow primarily in their own domestic currency.

(2) Their domestic currency can be created at will by their central bank, which these governments can control through legislation.

(3) These central banks have made no legal commitment to maintain the value of their currencies against any other currency or physical commodity – i.e. these are floating, fiat currencies.

(4) Therefore, if a government is ever unable to sell its bonds privately, it can always change the law and force the central bank to create new domestic currency to fund a deficit.

So sovereign currency issuers are always ‘solvent’, in terms of being able to make any debt and interest payments to bondholders in nominal terms, simply by forcing the central bank to buy up the new issuance required to fund these payments.

There is nothing profound about this. It just says that, if all else fails, you can monetise your deficit.

I cannot believe that this is put forward as a serious or practical policy argument.

I am all in favour of the ECB doing what’s needed to prevent continent-wide contagion from a Greek exit, including temporarily backstopping sovereigns. But governments should still do whatever it takes to avoid getting to the awful point of commandeering the central bank to avoid a certain default. And once there, they should do whatever it takes to get off monetary financing, as quickly as humanly possible.

Deflation can be brutal. Today’s Americans learn this from their country’s inter-war history.

Hyperinflation can be so much worse. Today’s Europeans learn this from Europe’s own inter-war history and its bloody, savage, senseless aftermath.

Yes, a sovereign currency can always turn to the printing press to prevent a default. And in extremis, once you’ve reached the point of ‘monetise or default’, monetisation may be the best of a bad set of options.

But don’t ever forget the horrors wrought on their citizens by governments that became addicted to the printing press.  And don’t be so damned blasé about it.

12 May 2012

Chart of the week: 7-13 May 2012

This week’s chart of the week (actually, charts) looks at the Bank of England’s decision to pause its government bond purchases (i.e. quantitative easing) at the current level of £325bn.

Ostensibly, policymakers are worried about prices. CPI inflation currently stands at 3.5%, above the Bank’s 2% target. But in spite of the Bank’s asset purchases, which have increased the monetary base by nearly 300% since March 2009, there is no evidence that this is feeding through to broader monetary aggregates. Indeed, M4 (broad money) has been shrinking since late 2010 – and at an accelerating rate!

Why the discrepancy between QE and falling M4? Because the money multiplier has collapsed, from more than 20 pre-crisis to 7.8 in March 2012. So all this extra monetary base the Bank is creating hasn’t fed through the broader money supply.

It’s therefore questionable that the QE measures up until now are inflationary. Rather, if policymakers want to affect broader monetary aggregates like M4, they’d have to go further up the risk spectrum than they’ve done so far, buying private assets that are less obvious substitutes for monetary base – and hence more likely to feed through to broad money.

10 May 2012

To QE or not to QE…

The Bank of England’s MPC decided today to leave the stock of quantitative easing at its current level of £325bn.

Gavyn Davies attributes their decision to a recognition that the supply side of the UK economy has underperfomed – suggested by the troubling fact that inflation expectations for 2012 have risen even as growth expectations have worsened:

A few thoughts on this:

1) The inflation-vs-growth chart is compelling, but I don’t fully buy the argument that we’re near the limits of what demand can do. The 2008-09 crisis may have had hysteresis effect on potential output, plus  pre-crisis growth rates may have reflected the bubble economy rather than being sustainable – but do we genuinely think all this is enough to leave real GDP today almost 13% lower than it would have been had growth continued on its pre-crisis trend? (Perhaps the answer is yes; but what’s the evidence?)

2) Even if you’re concerned that supply-side weaknesses could lead to more inflation – are the MPC really worried that QE is inflationary? Remember, the Bank of England’s asset purchases involving buying UK gilts – highly-liquid AAA-rated government bonds – or exactly the type of security that the shadow banking sector considers to be part of its (unmeasured) ‘reserves’ on the Singh/Stella basis. CPI inflation may be currently above the 2% target, but it’s hard to argue this a result of monetary factors. Indeed, broader measures of the money supply are falling in the UK: M4 growth in March 2012 was -4.5% yoy.

3) Finally, whilst it’s strictly outside the MPC’s mandate – could the UK not chose to tolerate a bit more inflation over the next few years than we were accustomed to pre-crisis? At least some above-trend nominal GDP growth seems to be a necessary condition for smoothing the process of public- and private-sector deleveraging over the next few years.

Overall then, a disappointing decision. We probably don’t just need more QE, but also purchases of a much wider range of assets by the central bank.

8 May 2012

Money multipliers and Basel III liquidity rules

Good Vox article by Manmohan Singh and Peter Stella about the transmission of monetary policy.

Textbooks tend to assume a very simplified model of the money creation process, in which central banks have direct control over the quantity of monetary base (i.e. currency plus commercial bank deposits at the central bank). The relationship between this monetary base and the total money supply is then governed by something called the ‘money multiplier’. This money multiplier relationship may arise due to legal minimum reserve requirements, which limit the amount of deposits a commercial bank can accept (and hence the amount of loans it can make) by the need to always hold the equivalent of X% of deposits in the form of reserves at the central bank – the quantity of which, as I said, the central bank directly controls.

I’ve always thought this model was odd. This is not least because, when I as an undergrad first looked in to what these ‘reserve requirements’ were, one of the first things I discovered is that we in the UK didn’t have a minimum reserve requirement – so in theory, banks could create an infinite amount of money from a given amount of monetary base.

Singh and Stella point out a further flaw. In today’s financial system, so-called ‘shadow banks’ (hedge funds, money-market funds, etc) clearly play a big role in the money creation process. But as well as not being subject to minimum reserve requirements, shadow banks also don’t have access to central bank deposit facilities. Therefore, for their liquidity needs, they have to rely much more on private sources – such as private repo markets. These typically require high-quality, highly-liquid assets (such as government bonds) as collateral, so shadow banks will hold such securities as their ‘reserves’.

Which leads on to Singh and Stella’s point – quantitative easing, in which the Fed and Bank of England bought up government bonds from banks, merely substituted ‘public’ bank reserves (i.e. new commercial bank deposits at the central bank) for ‘private’ reserves used by shadow banks (i.e. high-quality government bonds that were taken out of circulation). The true amount of monetary base was unchanged; so even for a given money multiplier, there will be no increase in the wider money supply.

And as a result, it’s incorrect to assume that the massive increase in measured monetary base since the crisis (Singh and Stella point out that deposits at the Fed increased from $20.8bn in 2007  to $1562.3bn in 2011) will be be inflationary – as there’s not a knock-on increase in the wider money supply.

Anyway, a (not-fully-formed!) thought follows on from this. The Basel 3 liquidity rules drawn up after the 2008-09 crisis will require banks to hold more liquid assets. I think these rules can be interpreted as a substitute for old-fashioned minimum reserve requirements, i.e. making sure that banks hold a sufficient chunk of liquid assets that enable them to meet deposit outflows. (Interestingly, the UK’s Financial Policy Committee recently left the door open to asking for powers to vary banks’ liquidity buffers over time, analogous to using minimum reserve requirements as a policy tool as is often the case in emerging market economies.)

But the Basel 3 liquidity rules’ definition of liquid assets will include both central bank reserves (i.e. Singh and Stella’s ‘public’ reserves) and high-quality sovereign debt (i.e. ‘private’ reserves). What does that mean for the money multiplier – in a world where Basel 3 supplants minimum reserve requirements as the binding constraint even for banks, does the concept of a conventional money multiplier relationship make even less sense?

%d bloggers like this: