Forward Guidance

What's it adding?

Last week the Bank of England, under new Governor Mark Carney, gave formal forward guidance on the future path of interest rates.  The Bank committed to keeping interest rates low until unemployment falls to 7.0 percent (it currently stands at 7.8 percent).  However, there were three caveats. The forward guidance would cease to hold if

·  in the MPC’s view, it is more likely than not, that CPI inflation 18 to 24 months ahead will be 0.5 percentage points or more above the 2% target; [or]

·  medium-term inflation expectations no longer remain sufficiently well anchored; [or]

·   the Financial Policy Committee (FPC) judges that the stance of monetary policy poses a significant threat to financial stability.

Markets behaved as if this was a tightening of policy, at least relative to what they expected (and such a move was widely anticipated).  Indeed, it is unclear how much this guidance adds to what we already knew.  The Bank's remit states that

In relation to monetary policy, the objectives of the Bank of England shall be:
  a.) to maintain price stability; and
  b.) subject to that, to support the economic policy of Her Majesty’s Government, including its objectives for growth and employment. [emphasis added]

Issuing guidance that monetary policy will remain loose while unemployment is high subject to inflation based caveats is not much more than a restatement of the above clause.  It certainly adds very little to George Osborne's recent update to the mandate, which emphasised that the inflation target is only the absolute priority in the medium term. Of course, the guidance gives some information to the markets about the bank's preferences by introducing the threshold of 2.5 percent.  But that is below even the 3.0 percent necessary for the Governor to have to write a letter to the Chancellor explaining what has gone wrong.  For all the fanfare, this watered down forward guidance has not added much.

A technical aside: this has been framed in the media as a commitment move.  But it is better seen as the Bank giving guidance on its preferences over output and inflation.  The classic Central Bank commitment problem is an inability to credibly promise higher-than-preferred future inflation, even when that would help push the real interest rate down to what is needed today.  However, the Bank is definitely not trying to promise future inflation - quite the contrary!

Vince Cable and "the capital Taliban"

The previous article was about equity ("capital") ratios in banks. Today, the Business Secretary Vince Cable has described those in the Bank of England pushing for more equity in banks as "the capital Taliban".  The FT's (freely accessible) Alphaville blog has an excellent list of problems with his remarks.

Bank regulation

Good headlines, insufficient policy
The government has today responded to the Parliamentary Commission on Banking Standards, which was set up to make recommendations on bank regulation in light of the LIBOR scandal. George Osborne is to implement most of the recommendations of the commission, notably the headline grabbing move to allow reckless bankers to be jailed.  This policy no doubt appeals to the public, and rightly so.  But it is far, far less important than requiring adequate equity (or 'capital') standards in banks - or in other words requiring a higher 'leverage ratio'.

The leverage ratio is the percentage of a bank which is funded by equity rather than debt.  It is so important because it determines how close to the edge a bank is operating.  A leverage ratio of, say, 10 percent means that a bank would be insolvent and in need of a bailout if it experienced a decline of 10 percent in the value of its assets.

On this crucial issue, the government has rejected the view of the Commission, which was also the view of the earlier Vickers Report, that the leverage ratio of 3 percent required by the most recent set of global regulatory rules ('Basel III') is too low.  Vickers had recommended a ratio of 4 percent.

Both of these numbers are far too low, and the government is wrong to endorse the smaller of them.  It is madness to have banks which cannot withstand a decline in asset values of only three percent. The principal argument offered by bank lobbyists and commentators against higher leverage ratios is that they reduce lending to the economy.  But a leverage ratio is a condition on how a bank is funded, not what it does with that money.  There is no good reason why higher bank equity should reduce lending.  In a recent book, finance academics Anat Admati and Martin Hellwig elegantly demonstrate that higher equity requirements make banks much safer, reduce perverse incentives, and come at no cost to society.   They refer to the arguments against them - such as falsely alleging that they will reduce lending - as articles of "the bankers' new clothes".

It is of crucial importance to society that banking is made safer, to avoid a repeat of the financial crisis.  This can be achieved at virtually zero cost to society. While headline grabbing measures like those announced today might cause some marginal improvements, the really important debate is being skirted.  Much bolder reform is needed.

Fiddling the figures on student fees

The Sutton Trust has abused its polling evidence

The Sutton Trust, an education charity, has today released some polling evidence on children's attitudes to the costs of higher education.  The charity claims that

Two thirds of 11-16 year olds have university finance fears...[and] worry about the cost of going to university

 This has been swallowed by the BBC, who report that

Two-thirds (65%) of 11 to 16-year-olds polled for the Sutton Trust voiced concerns about university costs

 A quick examination of the Sutton Trust's poll, at the bottom of its press release, shows that this is a distortion of the evidence.  The 'two-thirds' figure comes from the fact that roughly that proportion responded to the question

Thinking about the cost of going to university, which ONE of the following is the biggest concern for you? 

by selecting the prompted answers 'tuition fees', 'the cost of living as a student', or 'not being able to earn money while you're studying', as opposed to 'Does not apply – I’m not concerned about the cost of university', 'Don't know', or failing to tick a box.

This is an incredibly leading polling question. The charity has prompted children to think about the costs, and then asked them to pick the worst one. Imagine we polled on the following question:

Thinking about the risks involved in flying, which ONE of the following is of biggest concern to you?

With the options 'pilot error', 'engine failure',  'terrorism', and 'does not apply - I'm not concerned about the risks of flying'.  It is highly likely that a majority would pick one of the first three options; we've made them focus in on the risks and asked them to pick the worst one. I can easily identify the risk of terrorism as of 'more concern' to me than pilot error, even though I very rarely worry about either and neither would influence my decision to fly. What we have not done is asked whether our sample usually worries about the risks of flying, or how significant those worries are when it comes to their decision to fly.  If two-thirds selected one of the first three options, we would certainly not be justified in concluding that 'two-thirds worry about flying'.

This is irresponsible use of polling data from the charity, and poor journalism from the BBC who have reproduced the charity's dubious claim.  This is particularly damaging when it comes to student finance, as we face an uphill battle against poor understanding among the public of how the system actually works. This story has the potential to make children think "everyone else is worrying about student finance - that means I should", even though they certainly should not. That works against increasing access to university.

Finally, the Sutton Trust has advocated means testing tuition fees, despite the fact that they are currently paid only on the condition that you earn a significant salary.  This is bizarre; if, as they claim, children worry about fees despite this conditionality, what makes them think they will understand a complicated means-testing system any better?

Help to Buy

 The Chancellor is happy to increase debt, so long as it's not counted 

Last week was a bad one for George Osborne, Chancellor of the Exchequer.  The IMF called for him to reconsider the pace of his austerity programme. A numerical error emerged in an academic paper widely cited as part of the case for his economic policy.  Fitch became the second ratings agency to downgrade Britain's debt from AAA. And the Treasury Select Committee criticised the 'Help to Buy' scheme outlined in the budget.

Help to Buy involves the government lending money to people to buy houses. Banks typically ask for a deposit equal to about 25 percent of house value, which many struggle to provide. If you are buying a newly built home, the government will now step in and lend you 20 percent of the value, so that you only need a 5 percent deposit. This additional loan from the government will be interest free for five years.

Osborne continually insists that there is no room for debt funded fiscal stimulus on infrastructure investment, as some have called for. Yet, as Simon Wren-Lewis points out, Help to Buy is just this.  It is funded by more borrowing today, so the government can provide its twenty percent. Unfortunately, the corresponding investment is in very risky housing equity, which the state could easily make a loss on. The risks are increased by the fact that the lending will only be to people otherwise considered to have too little security.

Why has Osborne tolerated this? Because it is off balance sheet: the investment in the housing equity will not be counted in deficit figures. These statistics are what Osborne really cares about, as they affect his political narrative. But the policy undermines his argument: it is difficult to believe that investors in Britain's debt also care more about ONS press releases than about underlying fiscal obligations and risks.  According to Osborne, these investors won't tolerate more British debt - unless we don't count it.

Misleading cost comparisons

Thatcher funeral cost: poor journalism, poor argument
Whenever people oppose any item of state expenditure, they tend to make their point by comparing that expenditure to supposed alternatives. This week has seen many such comparisons due to £10m cost to the taxpayer of Margaret Thatcher's funeral.

Economists always urge using the idea of 'opportunity cost' - the options you have foregone - when assessing any expenditure. This highlights what you are implicitly choosing not to have when you spend resources on something.  Nonetheless, the way journalists carry out such analysis is almost always infuriatingly misleading and adds little to public policy debate.

Take this piece in The Guardian, which claims that the £10m to be spent on Margaret Thatcher's funeral could pay for, amongst other things, "322 nurses". The way they have calculated this figure is to divide £10m by a nurse's annual salary.  In other words, you would be able to employ 322 nurses for one year, and then make them all redundant.  They qualify other items of their list with time periods ("annual" water bills, "two years" of aid to Iraq, "10 days" of arts spending), but not the nurses, fire officers, or paramedics. This is probably because the image of hordes of potential extra service workers is too appealing to water down with qualifications.

Far better would be to take the Net Present Cost of employing one more nurse in the NHS from now on.  Assuming a two percent real interest rate, and using their salary figure, the Net Present Cost of employing one more NHS nurse from now on is £1.6m.  In other words, the opportunity cost of Thatcher's funeral cost is just over 6 nurses, rather than 322.

The reality is - whatever your view on Thatcher - a £10m one off capital expenditure is to the taxpayer essentially nothing. This is aptly illustrated by their jobseekers' allowance comparison.  They inform us we can pay for the long period of one week of unemployment benefits for less than a ninth of all claimants. Except we can't - they've used the lower 16-24 rate rather than the normal rate without telling us. Could it be that their 'Data Blog' - where 'facts are sacred' - is trying to push a certain view?  

Update: The Guardian has now changed the first three headlines to include the "for a year" qualification.

Bankers' bonuses

We should support performance related pay

Bankers' bonuses have dominated the headlines since the financial crisis.  Today brought two interesting stories.  Firstly, the EU has agreed in principle a deal to cap bankers' bonuses at two times salary.  Secondly, RBS  (which is 82% owned by the taxpayer) announced it lost £5.17bn in the last year yet will pay out bonuses totalling roughly £600m, with £210m of that going to investment bankers.¹

No doubt, the move by the EU will cause cheer in the popular press.  However, it has been widely documented that as curbs on bonus payments have been introduced, banks have increased base salaries instead.  For any given pay package, taking more of it as salary and less as bonus makes a banker better off, as they face less uncertainty about their (very high) pay.  It also means that poor performing bankers cannot be so easily punished by lower-than-expected pay.

The public finds the 'bonus' word so toxic in the case of losses because they assume a bonus must be a  reward for doing well.  Why could it make sense to have any reward when a bank makes a loss?  In reality, the city operates on the basis that employees loosely expect a bonus of a given size, which is increased when there is good performance, and decreased when there is bad performance.  So a positive bonus can still translate into a punishment for poor performance, if it is below expectation.

This regulation reduces banks' ability to operate under such a performance-related system.  There may be good reasons for that, if bonuses are incentivising traders to take harmful risks.  Nonetheless, the regulation has the potential to make bankers better off by giving them more no matter what happens - and this is not a point you will find discussed much in the press.

In the RBS case, the devil is in the detail.  Most of the loss came from an accounting charge of £4.65bn due to the "fluctuating value of its own debt and derivative liabilities", according to the FT.  The operating profit was £3.46bn, up from £1.82bn a year ago.  And guess what?  Markets & International Banking - the investment banking arm - contributed £2.1bn to that.  Do bonuses totalling £210m still look so unreasonable?

An aside:  it always bothers me how bonuses are reported on a total rather than per-employee basis.  For those who are interested, the RBS results show 15,600 employees working in the Markets and International Banking divisions. Assuming the BBC's quoted bonus figure of £210m for investment bankers (see footnote 1), that works out at an average bonus of roughly £13,500.

1. I have taken the number for the bonus payout for investment bankers from the BBC.  However, my numbers on operating profits and headcount are from the RBS annual statements.  I am assuming the "investment banking" arm to be the International Banking and Markets divisions combined, but it is possible that the BBC has used a different definition.