Thursday, May 19, 2011

Much applied work in economics concerns the problem of which came first: chicken or egg. In a recent paper, Michael Hilmer and Christina Hilmer consider this problem as it applies to the context of PhD students. They investigate the early success of PhD graduates in publishing their work, and ask whether this is due to the eminence of their supervisor - or rather is it the case that supervisors with good reputations tend to attract intrinsically more able students?

In this example, the direction of causality can be inferred by examining the relative performance of students studying with different supervisors in the same doctoral programme, and by examining the relative performance of students who share the same supervisor where the supervisor works in more than one programme (usually in different institutions).

The results suggest that success is due more to the individuals' innate abilities than to the supervisors. Good supervisors tend to attract good students, but the early career publishing performance of a given supervisor's students is higher in doctoral programmes with a higher quality (that is, in programmes which likely attract more able students).

For universities, therefore, getting the programme right - attracting the right students - is a clear prerequisite for success. Understanding which comes first - chicken or egg - is crucial.

More broadly, the chicken and egg question is key to understanding a wide range of policy issues. Which causes which: class size or scholastic performance? risky behaviours of depression? investment or output?



Hilmer, M., & Hilmer, C. (2011). Is it Where You Go or Who You Know? On the Relationship between Students, Ph.D. Program Quality, Dissertation Advisor Prominence, and Early Career Publishing Success Economics of Education Review DOI: 10.1016/j.econedurev.2011.04.013

Wednesday, April 27, 2011

I have occasionally on this blog referred to a neural network forecasting model of the UK economy that I like to use. I most recently reported on forecasts from this model in January of this year. At that time, the model was predicting a fall-off in growth towards the end of summer of this year. The most recent experiments that I have run with this model use data up to February 2011, and they still indicate that growth will falter towards the end of summer.

The caveats surrounding this model are many: it's based on industrial production data (since this is available monthly), and doesn't include information about other variables - notably policy. But for a simple model it has a rather good track record. While the pessimism in the model's forecasts may not be widely shared, these figures do serve to emphasise just how fragile is the recovery.

The graph below shows the actual data back to just before the downturn - the forecasts are at the right of the graph (where the line turns red) and look forward 18 months.

The growth figures announced today indicate that the UK's economy grew by 0.5% over the first 3 months of this year. In the last 3 months of last year, it fell by 0.5%. So we are not yet quite back to where we were in the third quarter of 2010. (A fall of 5% followed by a rise of 5% nets out as a fall of 0.25%, not zero, because the second 5% is calculated from a lower base.) The contraction in the last quarter of 2010 was, we were told, due to poor weather. If that were really the case (and the evidence suggests that, in part at least, it was), then the improvement in the first quarter of this year is surely due to the absence of poor weather - and nothing more fundamental than that. As the Chancellor of the Exchequer said yesterday, 'we are not out of the woods yet'; the recovery, such as it is, remains very fragile.

That said, from this low base, there are now some signs that offer encouragement. It remains essential for the government, in its zeal to attack the budget deficit, to pay heed to the impact that its policies has on growth.

Thursday, April 21, 2011

Are we seeing some signs of life? Consumer confidence up,retail sales up, unemployment down...

One would hope so. But the improvement in these statistics is slight, and in each case the data follow on from very bad figures in the previous month. If indeed we are seeing the first indications of recovery, that recovery remains very fragile. Whatever one's views might be about the speed with which the budget deficit is being cut, the policy shock that still has to be played out is likely to result in a bumpy ride over the coming months.

Wednesday, April 06, 2011

Many observers have noted that the financial situation facing the UK is different from that in Greece, Ireland, or Portugal. The rates of interest which the latter countries have had to offer in order to sell their Treasury bills illustrate this. While six month T bills can be sold by the UK at an interest rate of around 0.7%, Portugal has today auctioned off six month T bills at a mammoth 5.1%. This figure makes the terms of any prospective bail-out attractive to the Portuguese authorities, and such a bail-out is now surely inevitable.

Wednesday, March 23, 2011

The revisions made to the Office for Budgetary Responsibility's growth projections, released to coincide with today's Budget Statement, are in a downward direction, with 2011 growth now anticipated to be 1.7%. This still looks a touch optimistic.

The Chancellor of the Exchequer described his Budget as a Budget to fuel growth, but beneath the hype there does not appear to be much fuel in the tank. To some extent this is unsurprising as the Chancellor has already used the Comprehensive Spending Review to tie his own hands. Even so, further - and more accelerated - redistribution towards those on lower incomes would have done at least something to improve the multiplier effect of what measures are in place to support the economy. The £600 rise in the personal allowance on income taxation does not kick in until April of next year - too little and too late to help an economy that is shrinking now. And a shrinking economy makes it all the more difficult for the government to meet its targets with the public finances.

The Budget has been accompanied by something called a 'Plan for Growth'. This aims to increase the competitiveness of the tax system, to encourage the creation and development of businesses, to rebalance the economy through investment and export activity, and to provide a more educated workforce. The headline ambitions suggest that the government's aim is to produce a raft of microeconomic policies that could help improve competitiveness and hence provide the right context within which economic growth can take place. So far, so worthy.

When we turn to the detail of the Plan for Growth, remarkably little involves a financial commitment by the government. There is reference to an additional £100 million to repair potholes, and one or two other things. But the (uncosted) words 'encourage', 'support', and references to reducing 'burdens' occur frequently. In other words, it's a 'do it yourself' plan for growth - much as the expenditure cuts were a 'do it yourself' exercise (a huge proportion of these cuts being shifted onto local authorities). The plan - inasmuch as it is a plan rather than a hope - is for a Thatcherian resurgence in enterpreneurship. We should wish for this to be successful. Whether or not it will be so will depend in large measure on the detail of the microeconomic policies that are introduced. And for that, we'll have to wait and see.

Monday, March 21, 2011

This week's budget needs to include policies aimed at the promotion of growth. There is a wide range of options open to the government, but one that is in line with the policies of one of the coalition partners (the Liberal Democrats) would be to raise the personal allowance (or 'disregard') on income tax. For the coming year this is set to be £7475. A substantial increase in this would raise disposable incomes - especially those of the people on the lowest incomes. Since the latter group tend to spend the largest proportion of their incomes, the knock-on ('multiplier') effects of raising their disposable income is likely to be greater than that associated with raising the incomes of other workers.

Ultimately the Liberal Democrats want to raise the personal allowance to £10000. While their more hawkish partners in government are unlikely to want to go the whole hog at this stage, it is, of course, now that the stimulus is most needed. Tax revenues fall when the economy contracts, and - while borrowing fell over most of last year while the economy was starting to recover - the more recent stagnation is doing nothing to help the public finances. Moving most of the way to the £10000 target - in a budget containing other policies to encourage growth - would be prudent.

Friday, March 18, 2011

Consumer confidence has, according to the Nationwide's monthly survey, hit an all time low. What is particularly worrying about these figures is that consumer confidence has a track record as a very good leading indicator of the state of the economy. Confidence has clearly not been helped by the government's failure, thus far, to deliver any clear plan for growth. While the hope is often expressed that the private sector will expand to soak up the supply of labour that is shed by public organisations, this has, so far, been just that - a hope. The reality has been one of rising unemployment - with the number unemployed rising to 2.53 million in January, the highest figure for some 17 years. Meanwhile, the OECD has provided a downbeat assessment of the UK's growth prospects over the next 2 years. This anticipates growth of just 1.5 per cent this year. The OECD does deem fiscal consolidation to be 'necessary', but also describes the government's plans in this area as 'ambitious'.

Next week's budget needs to provide what has heretofore been lacking - not just a hope, not just a promise, but robust government support for growth. And if that means trimming the ambition just a little, and losing face just a little, then so be it.

Tuesday, March 15, 2011

There are some very pertinent observations about the future development of macroeconomics in this talk by Olivier Blanchard.

His nine points are:

1. In the wake of the crisis, policy making has to change, in particular with less of a focus on a single policy goal.
2. The pendulum has swung, some at least, in the direction of a recognition of an enhanced role for the state in macroeconomic regulation.
3. We previously underestimated the impact on the macroeconomy of microeconomic distortions.
4. There exists a multiplicity of macroeconomic policy tools...
5. .... not all of which are fully understood.
6. Some of these tools are difficult to employ for reasons of political economy.
7. The future research agenda for macroeconomics, with solid microfoundations, is exciting.
8. There should exist a multiplicity of policy goals. But moving from a model of inflation targeting to one where multiple goals are targeted is difficult.
9. We should not hope for too much.

Recent developments in macroeconomics have focused on models based on dynamic stochastic general equilibrium (DSGE). These models are characterised by a set of macroeconomic relationships which are built up from consistent microeconomic foundations. Their microeconomic underpinnings are intended to finesse the criticism raised by Lucas that changes in policy beget changes in the behaviour of microeconomic agents thereby posing a challenge to economic forecasters. DSGE models come in a variety of flavours - there are classical models developed from the real business cycle literature and there are Keynesian models which build in price rigidity as an institutional feature. The models take the assumption of rational expectations as given. Large scale variants of these models are used as a tool for policy making and evaluation in central banks.

In recent work, Georg Weizsäcker has produced compelling evidence that - contrary to much of the economic orthodoxy of the last 40 years - people do not form their expectations rationally. It takes particularly strong evidence to disturb someone's expectation away from their prior beliefs. If the expectations formed by an agent are approximately correct, they will not adjust them rationally. Only if they are way out will they take new evidence on board.

Weizsäcker's findings should not be particularly surprising. They suggest that a way forward in modelling may be to adopt a nonlinear approach wherein the degree of rationality used in forming expectations varies inversely with the accuracy of those expectations. Given the evidence, it is clear that our models would benefit by admitting some sort of learning mechanism - something more refined than the assumption that we always achieve rationality.

Weizsäcker, G. (2010). Do We Follow Others when We Should? A Simple Test of Rational Expectations American Economic Review, 100 (5), 2340-2360 DOI: 10.1257/aer.100.5.2340

Wednesday, March 09, 2011

In recent work, Atish Ghosh, Enrique Mendoza and Jonathan Ostry have developed a means of evaluating the 'fiscal space' of an economy. This is a measure of the flexibiltity that the authorities have to engage in expansionary fiscal policies given the levels of national debt. The measure is calculated by reference to the maximum level of national debt that (given that interest payments need to be made on the debt) is sustainable at the economy's GDP. If the national debt were to rise above this point, then interest payments will serve to ensure that it would rise forever more, and so default would become inevitable.

The results are instructive. Unsurprisingly, Iceland, Greece and Portugal are amongst the countries with very little headroom. Italy and Japan are also in this group. Spain and Ireland (surprisingly, perhaps), are somewhat better placed. The UK is (perhaps also surprisingly) in a relatively comfortable zone, keeping company with countries such as France and Germany.

Thursday, March 03, 2011

Results from the new panel dataset, Understanding Society, have been reported today, suggesting that people in the UK remained broadly optimistic at the height of the recession in 2009. At times like these, good news is always welcome. It is important, however, to note that this is a snapshot. Evidence from another source - the Nationwide survey of consumer confidence - confirms that confidence did indeed recover well during 2009. Indeed by the end of that year, confidence was not far below the levels realised in the summer of 2007, before the first signs of the credit crunch became fully apparent. However, since the end of 2009, confidence has plummeted once more and the Nationwide's index is currently close to the minimum value that it realised at the end of 2008.

The world moves fast. Understanding Society is set to be a fantastic source of data for researchers, but when analysing data we should pick horses for courses. Understanding Society is simply not up to date enough to provide useful information about consumer sentiment now.

Thursday, February 17, 2011

In a post on this blog last June, I expressed concerns that the Office for Budgetary Responsibility was underestimating the magnitude of the output gap in the UK. This underestimation appears to be contagious. A member of the Bank of England's Monetary Policy Committee, Andrew Sentance, has argued in a speech that the output gap is low and that maintaining historically low interest rates is stimulating the economy beyond the point at which inflation can be held in check.

The notion that, with high and rising unemployment, the output gap is low seems to be difficult to defend. There are, to be sure, reasons why prices are rising. Oil and food prices have risen internationally for reasons that have been well documented (here and elsewhere) and which are likely to be transient. Tax hikes have also contributed.

The economy shrank in the last quarter. Fiscal policy is imposing a squeeze which will inevitably increase the challenges faced in the labour market. To move too early to accompany this with a monetary squeeze - and to do so on the basis of price statistics that are probably blipping - would be dangerous. Yes, we need to keep an eye on inflation, but, no, the price increases we have seen up to this point do not suggest that inflation is a significant threat.

Wednesday, February 09, 2011

Project Merlin - the initiative that has aimed at securing more bank lending to businesses by linking bankers' bonus payments to such lending - has finally, after much wrangling, delivered an agreement. The banks will raise their lending to a total of around £190 billion this year. Some 40% of this will go to small firms. The banks will also provide capital (some £200 million) for the Big Society Bank - which is intended to fund projects that are community based. There will be increased transparency in bankers' pay, but this falls short of the hope that bonuses would be contingent on banks meeting lending targets.

Lending will be monitored by the Bank of England which will provide regular reports - which presumably means naming and shaming banks that do not live up to their promises. Beyond this, it is not clear that Merlin has any teeth. Banks will lend, and they will do so at rates of interest that they themselves set. If businesses do not find these rates attractive, they will not look for loans.

In this context, it is a little disquieting to note the continued rise of LIBOR - which is now at 0.80%, some 30 points above the Bank of England's interest rate. There is still a lot of nervousness around the financial markets, and that means that lending is still not going to come cheap.

So the environment in which Merlin has been launched is not altogether promising. A little more prescription might not have gone amiss.

Wednesday, January 26, 2011

The big question to arise from yesterday's disappointing growth figures is: can it really all have been about the snow?

Occasionally over recent months, I have referred to a forecasting model that I like to use. If the slump in the last quarter of last year was indeed due to the exceptional weather, then this model should fail to predict it. Indeed my model fails to predict the final quarter slump. So snow may have played a part. (Bear in mind, though, that this is a loose test - not least because the data I use in this model are industrial production data, and we know that manufacturing did relatively well in the last quarter.)

Despite this, the model's forecasts are not all that encouraging. They are that production will show modest growth (at an annualised rate of less than 2%) up to the late summer - but then it starts to fall off.

Plenty of caveats attach to a simple model of this kind (a neural net with a hyperbolic tangent squasher, 24 months of input data, 2 hidden layers, where the data are the 12-month change in logged index of industrial production data), or indeed to any economic forecast. But, for a simple model, the track record of this forecaster is pretty good. Snow may have played its part in the last quarter of 2010, but the recovery is nonetheless far from secured.

Tuesday, January 25, 2011

It's a double dip. The 0.5% contraction in the economy over the last quarter of last year represents a truly awful outcome. The government has been quick to blame poor weather towards the end of December. But that is clutching at straws.

The figures are considerably worse than the expected increase in output of around 0.4%. It represents a dreadfully weak platform on which to build the government's austerity measures. It is time for a reassessment.

Thursday, January 20, 2011

Figures released this week indicate an increase in the unemployment rate to 7.9% and an increase in the inflation rate to 3.7%. Both are a concern, and they bring to mind the 'stagflation' that was experienced in the 1970s.

Economics is often thought of as the 'dismal science' - when there is good news about unemployment, there is usually some bad news about inflation, and vice versa. When there is bad news about both, things get really dismal, and we really have to try hard to look for some good news. There may be some. Petrol prices have risen sharply; food prices have also increased. The price of oil does tend to fluctuate a lot (a fact not unrelated to the inelastic nature of both demand and supply in this market), and at close to $100 per barrel the price is now, if anything, above its steady state. The increase in the price of food has been driven largely by severe, but transient, weather events.

So where is the good news? As things stand, it looks as though inflation is blipping. It will rise further before falling - not least because of the effects of the January VAT increase. But as oil and food prices stabilise, the inflationary pressure should ease.

This does suppose that the inflation that we have experienced up to now will not feed through into wage increases. If it does, then stagflation indeed looms. The Bank of England's Monetary Policy Committee is sure to look closely at wage settlements over the coming months. If there were to be a general upward drift in wage settlements, then that would not be good news for inflation - wage hikes would feed through into further price hikes, and this would lead to wages and prices leapfrogging each other in an inflationary spiral. To avoid this, the Bank may need to raise the interest rate sooner rather than later - and that would further threaten the recovery. Let us hope that sense prevails.

Monday, January 10, 2011

LIBOR has been creeping up over the last quarter or so, and has now risen to 27 points above the Bank of England's rate. All is still not well in financial services.

Monday, December 20, 2010

The Conferderation of British Industry has lowered its growth forecast for the coming quarter; it now forecasts quarter-on-quarter growth of just 0.2 per cent over the period from January to March 2011. It does, however, still expect growth of 2 per cent over the coming year.

Meanwhile, the Nationwide consumer confidence index fell further in November, and is now close to the low point that it reached in January 2009. (The index stands at 45 in November; it was 43 in January of last year.)

I have, on a couple of occasions over the last few months, reported the predictions of a neural network forecasting model. Over the coming 18 months, that model is currently predicting modest growth for the UK, and suggests that the danger of a double-dip is receeding. It should, however, be borne in mind that this model is based on output data (monthly data on industrial production, to be precise); the model may not adequately allow for the threats to the economy that come from sluggish consumer spending, fragile overseas demand, and government austerity packages. Nonetheless, the model offers a little festive cheer.

Thursday, December 09, 2010

The Institute for Fiscal Studies has reported on the government's proposals for student finance. Its findings cast some useful light on a debate that has been dominated by political spin. An important conclusion is that 'graduates from the poorest 30% of households would (under the government's proposals) pay back less than under Lord Browne's proposed system, but more than under the current system'. Many politicians would have you believe otherwise.

Browne's proposals, which involved a 'soft' cap on tuition fees at £6000, with universities being allowed to raise fees above that level while being subjected to a steeply rising tax, have been replaced by a 'two cap' system - where the cap is either £6000 or £9000 conditional on widening participation. Both Browne and government proposals involve a loan for tuition fees and maintenance that is repaid once the student graduates and is earning £21000 or more per year. Concessions made by the government include the annual uprating of this £21000 figure, and tuition fee discounts to students from the poorest families.

The media has portrayed this as a three-fold increase in tuition fees, indicating an expectation that many universities will pitch their fees at or near the higher cap. Indeed, it is quite possible that, by introducing the higher cap, the government will already have encouraged some institutions to set fees higher than they would otherwise have chosen. This would appear to be good for neither students not public finances, and appears to have been driven by a political need to retain a cap of some sort. Yet surely it is almost inevitable that the cap will go some day, and Browne's suggestion of protecting students and the taxpayer by imposing taxes on universities that charge high tuition fees was eminently sensible. In this respect, politics seems to have trumped common sense.

The universities, meanwhile, will suffer severe cuts to their teaching budgets from the coming tax year (which starts before the end of the current academic year), and will - if the legislation goes through - be able to charge higher tuition fees only from academic year 2012-13. They will face a more robust competitive environment, and will need to innovate rapidly - and constantly - if they are to prosper. These are interesting times indeed in higher education.