Tag Archives: macroeconomics

How not to do economic forecasts

Posted on by

Every so often I come across a book that should be read by: (a) all economists; (b) all students; (c) everybody involved in politics and policy; and (d) everybody else with an interest in the world. Nate Silver’s [amazon_link id=”0141975652″ target=”_blank” ]The Signal and The Noise: The Art And Science of Prediction[/amazon_link] – which I’ve finally read shamefully long after publication – is one of those books. It should in fact be read by all macroeconomists who publish forecasts at least annually, as a condition of their continuing membership of the profession. If I were teaching, it would emphatically be a required item on the course reading list.

It is a wonderfully clear and engaging explanation of the challenges of making predictions in fields ranging from politics and elections to weather and earthquakes to economics and poker. Apart from a couple of sections on American sports, which might as well have been written in a foreign language, the examples illustrate how to, and how not to, make forecasts. You’ll be wiser for reading it, not to mention able to put Bayesian inference into practice. Silver makes a compelling case for adopting the Bayesian approach, rather than the standard (‘frequentist’) statistics descended from R.A.Fischer and universally taught to economists in their econometrics courses. The emerging new economics curricula should at least include Bayesian statistics in the modules covering empirical methods. As Silver writes:

“Essentially the frequentist approach toward statistics seeks to wash its hands of the reason that predictions most often go wrong: human error. It views uncertainty as something intrinsic to the experiment rather than something intrinsic to our ability to understand the real world.”

In other words, it is not true that collecting more and more data – although usually useful to a forecaster – will eliminate your uncertainty about the real world. The signal-noise problem is epistemologically unavoidable. What’s more the frequentist approach involves assumptions about the distribution of the population; we know about the (in-)validity of the normal curve assumption, and anyway, “What ‘sample population’ was the September 11 attack drawn from?”

The chapter on macroeconomic forecasting is a pretty devastating critique of economists who do that kind of thing. There is a demand for macro forecasts, and I’d rather economists supply them than anybody else. But we shouldn’t pretend they’re going to be accurate. Almost all forecasters, even if they publish standard errors, will give the impression of precision – is growth going to be 0.5% or 0.6%? – but it is inaccurate precision. Silver calculates that over the period 1993-2010, GDP growth fell outside the 90% confidence intervals of macro forecasts for the US economy a third of the time, and a half the time if you look back to 1968.

Macroeconomic data are very noisy, especially early estimates of GDP: in the US the margin of error on the initial quarterly estimate of GDP is plus or minus 4.3%. The initial estimate for the final quarter of 2008 was a decline of 3.8% – later revised to minus 9 per cent. Silver makes the comparison between economic forecasts and weather forecasts, similarly difficult problems. However, weather forecasting has improved over the decades, thanks to a better understanding of the causal links and a greater degree of disaggregation of data, made possible by more powerful computers. Economists have neither the improved understanding – on the contrary, important causal links notably finance were ignored until recently – not seemingly the appetite for better data (as I’ve pointed out before).

The book also makes the point that others (like [amazon_link id=”0465053564″ target=”_blank” ]Paul Ormerod[/amazon_link]) have emphasised, that the economy is a complex non-linear system so there is a lot of unavoidable uncertainty about forecasts more than a short period ahead. It also notes that although we know about the Lucas Critique and Goodhart’s Law – both pointing out that policy affects behaviour – economic forecasters typically ignore it in practice. Silver also underlines the rarely-resisted temptation to overfit the data – and microeconomists are just as guilty as macroeconomists here. The temptation is strong because an over-fitted model will seem to ‘explain’ more than a ‘true’ model when the data are noisy, so the usual tests for good fit will look better. [amazon_link id=”0472050079″ target=”_blank” ]Deirdre McCloskey and Stephen Ziliak[/amazon_link] have been pointing out the siren allure of ‘statistical significance’ for ages – it has almost nothing to do with economic meaning – and perhaps The Signal and the Noise will help broadcast the message further.

Finally, I learned a lot from the book. The chapter on how to approach the question of CO2 emissions and climate change is a model of clear thinking. My favourite new fact: members of Congress – with access to lots of company information via lobbyists and the ability to influence companies’ fortunes by legislation – see a profit on their investments that beats the market averages by 5 to 10 per cent a year, “a remarkable rate that would make even Bernie Madoff blush,” as Silver observes.

Anyway, if you haven’t yet read this, go and do so now. The new UK paperback also has a wonderful cover image!

[amazon_image id=”B0097JYVAU” link=”true” target=”_blank” size=”medium” ]The Signal and the Noise: The Art and Science of Prediction[/amazon_image]

Update: Dan Davies (@dsquareddigest) has gently rebuked me for the paragraph about Bayesian versus frequentist statistics. Via Twitter, he said: “Silver has a really annoying misintepretation of Bayesian vs frequentist which is now becoming commonplace… the paragraph you quote is really confused – NS is a good practical statistician but all over the place on theory & methodology. The (otherwise excellent) book gains nothing from taking a very strong position in someone else’s philosophical debate.” Needlesss to say, I know less than Dan about this debate. This doesn’t change my mind that econ students should be taught the Bayesian approach too, nor the conclusion that the book clearly explains how to do it in practice.

Macroeconomics – is it all underpants?

Posted on by

A couple of days ago, Simon Jack of BBC Radio 4’s Today Programme interviewed me about unconventional economic indicators. We chatted about the cranes index, written up by Chris Giles in the FT as a marker of regional imbalance in the UK economy, about hemlines, lipstick, champagne sales. One new to me, unearthed by the researcher, was Alan Greenspan’s supposed interest in sales of men’s underpants. I thought it was in Greenspan’s book, [amazon_link id=”0141029919″ target=”_blank” ]The Age of Turbulence[/amazon_link], where he does talk about his interest in detailed economic statistics, but it turns out the source is a 2008 NPR interview about the book.

[amazon_image id=”0713999829″ link=”true” target=”_blank” size=”medium” ]The Age of Turbulence: Adventures in a New World[/amazon_image]

My discussion on the radio was mildly frivolous, but the light-heartedness covers a serious point about macroeconomics, namely how studiously unempirical it is. This might seem a contrarian statement, given how frequently macroeconomists bandy about debt-GDP ratios, GDP growth rates, inflation and unemployment rates. But in fact Greenspan was something of an exception with his obsession with the statistics underpinning the aggregates. Most macroeconomists trade blows with the same aggregate figures drawn from the same online databases, and their differences are disagreements about the interpretation of the figures in the light of their prior beliefs about a ‘true’ model of the economy. They demonstrate confirmation bias in finding aggregate figures to support their views.

One of the problems with macroeconomics, therefore, is how little attention its practitioners pay to either understanding the construction and intellectual framework underpinning the aggregate statistics they do use (none of them being natural entities, all analytic constructs – see my forthcoming GDP book), or to collecting new statistics. So I’m with Alan Greenspan on this point, and think sales of underpants could be more revealing than the conventional figures.

Austerians, Stimulards, Krugmanites and history

Posted on by

If you divide history up into separate decades or eras – arbitrary, of course – the average growth rate of similar economies differs greatly between them. For example, growth was 1-3 percentage points slower in all of the major western economies in 1973-1995 as compared with 1950-1973. If this doesn’t sound much, remember the power of cumulative arithmetic: at 2% a year growth, real incomes double after 34 years, compared to just 23 years at 3% a year growth. Stephen King starts his terrific new book [amazon_link id=”0300190522″ target=”_blank” ]When The Money Runs Out: The End of Western Affluence[/amazon_link] by suggesting that the west is currently in one of the eras of structural slow growth – and goes on to argue that there are good reasons to expect this to last for a long time. So he also starts out by picking an argument with all those economists and commentators who present the debate as a cyclical one, i.e. asking how can policymakers correct for the downturn and get things back to trend? King is neither an Austerian nor a Stimulard: “Both sides believe in economic recovery. Each happens to think that the opposing view is totally wrong.”

[amazon_image id=”0300190522″ link=”true” target=”_blank” size=”medium” ]When the Money Runs Out: The End of Western Affluence[/amazon_image]

This gets the book off to a good start, as far as I’m concerned. Whenever I’ve voiced, far more tentatively than it does, some consternation at the current macro debate, I’ve been shouted down by people on each side who tell me that I’m just wrong, and the data prove for a fact that their view is correct. What’s more, certainty is popular – wouldn’t we all love things to get back to the way they were pre-crisis?

A second point the book makes early on is that, precisely because of the stagnation, “Economic policy is no longer for the technocrats. It has become inherently political.” Again, I wholly agree. Structural slowdowns in growth will not end without structural economic reforms, and that’s economics jargon for difficult political choices. The historical episodes described in the book, some well-known, others less so, help shed light on the type of political dilemmas facing western economies now.

The first chapter looks at the roots of the current stagnation, and finds them in the common presumption that economic growth could be taken for granted. Some of the examples are staggering – for example, I learned that by the end of the 1980s it was not uncommon for Japanese homebuyers to take out 100 year mortgages, thus explicitly living on their children’s incomes. In all the western economies, future generations have been defrauded in more and less overt ways – and again, I wholeheartedly agree (this was a theme of [amazon_link id=”0691156298″ target=”_blank” ]The Economics of Enough[/amazon_link]). The debt overhang consists not just of financial instruments but also political promises that might not be achievable.

The book’s subsequent chapters set the policy response to the current crisis alongside a number of historical examples. King notes that the large economic stimulus, mainly through monetary policy, has meant growth post-2008 hasn’t been as bad as it might otherwise have been. But he’s sceptical about ongoing quantitative easing on the present massive scale. “If QE fails to deliver a lasting recovery in economic activity, it shifts from being part of the solution to becoming part of the problem.” And he argues that the impact of QE on growth is unpredictable, with a larger impact on the distribution of economic activity than on its level.

King picks a particular argument with Paul Krugman, which (I know from my own modest experience of mildly criticising Krugman’s messianic certainty) will bring much ire down on his head. He believes Krugman is overly obsessed with parallels between the present and the 1930s ([amazon_link id=”0393345084″ target=”_blank” ]End This Depression Now![/amazon_link]), overlooking some important differences. King points out that the value of national income in the US declined in the 1930s – there was deflation. Now, the volume of US GDP has fallen short of expectations, but the value has not, and fears of deflation proved unfounded. No doubt the Krugmanites would explain that this is due to the fact of stimulus policies. The book’s counterargument is that inflation has run ahead of expectations for some years, pre-dating the crisis, and is due to a progressive deterioration of the economy’s supply potential. “In any case, the ammunition available to Roosevelt no longer exists,” he adds; FDR inherited a healthy fiscal position, and under him the budget deficit peaked at 9% of GDP, in contrast to the large pre-crisis deficits in the US and many other countries.

King’s conclusions are gloomy – the title of the penultimate chapter is ‘Dystopia’. Trust in banks, politicians, foreigners, business and more has declined. There are strains between haves and have-nots, between old and young, between regions. (I’ve written about declining trust in an essay for the OECD ahead of its annual Forum later this month.) There is an entitlement culture, including elites like bankers, which prevents fiscal reform. Globalization may be going into reverse. We are far from hyperinflation, but higher inflation can co-exist with stagnation, as it did in the 1970s. Political extremism may be on the rise.

Can anything be done? A brief final chapter is called ‘Avoiding Dystopia’. The key challenges are addressing the global savings imbalances that lay at the root of the crisis, creating something close enough to fiscal union for the Eurozone to work, bring down high levels of government debt over time with a lasting and credible commitment to lower government spending relative to revenues, find mechanisms to take account of the interests of future generations (in ageing societies where pensioners are the most likely to vote), and unwind QE and put in place a new and credible monetary framework such as nominal GDP targeting. Oh, and sort out the banks’ balance sheets and regulatory regime, fix the education system, and reform the economics profession. If anything, seeing written down the scale all of these challenges is even more depressing than the ‘Dystopia’ chapter.

I’ve got no doubt that even my setting down a description of a book that isn’t avidly anti-austerity will bring down on this post the wrath of the Krugmanites and Stimulards. But I’d urge both teams, both Austerians and Stimulards, to be a tiny bit open-minded and read the book. Look at the past history of growth: there is no guarantee that it must recover to 2% or more a year. Is it not possible that it’s more important for policymakers to address the underlying structural challenges? I fear that unless attention moves away from the Punch and Judy act of so much macro debate, we’re bound to face a long era of economic stagnation.

 

What is different, and what isn’t

Posted on by

I’ve been a bit mystified by Excel-gate (see this good, balanced summary by Gavyn Davies). Bravo for Thomas Herndon, the graduate student who uncovered the error in the now-notorious paper by Carmen Reinhardt and Kenneth Rogoff; his job prospects will be rightly enhanced by this episode.

But the glee with which anti-Austerians pounced on this episode to ‘prove’ that austerity doesn’t work seems to involve an assumption that the original Reinhardt-Rogoff paper of 2010 ‘proved’ anything to the contrary in the first place. There are lots of papers about the impact of debt/GDP ratios on growth, and they demonstrate all kinds of different things – see for example this BIS paper by Cecchetti and others, or this IMF paper (pdf) from last year on the Caribbean economies, or this Fed paper published in December (pdf), or this much-cited 2010 paper by Koehler-Geib and others, or for that matter the new paper debunking Reinhardt and Rogoff’s 90% as it too finds the same correlation albeit with different numbers.

Well, you get the idea. Taking these together, we ‘know’ there might be a threshold for sovereign debt, but it varies over time and across countries, it’s a correlation whose causal direction and mechanism is unclear, and there isn’t enough data for any estimates to be robust (because history only runs once). All of which only goes to underline how little is known about the macroeconomy, not to mention how hard any macroeconomists and their camp followers find it to resist claiming certainty where there is none.

No doubt Reinhardt and Rogoff were tempted into over-claiming for their work by the politicisation of the debt threshold issue. But the underlying message of their big 2009 book, [amazon_link id=”0691152640″ target=”_blank” ]This Time is Different[/amazon_link], is unscathed: unlike the later paper, it makes it absolutely clear that debt ‘thresholds’ above which increasing borrowing is correlated with lower growth vary widely in different countries and at different times (no magic 90% here); and that the historical record indicates it generally takes a long time for growth to recover after banking crises involving debt overhangs.

[amazon_image id=”0691152640″ link=”true” target=”_blank” size=”medium” ]This Time Is Different: Eight Centuries of Financial Folly[/amazon_image]

Austerity and the barbarian horde

Posted on by

Here’s a book that does what it says on the cover: [amazon_link id=”019982830X” target=”_blank” ]Austerity: The History of A Dangerous Idea[/amazon_link] by Mark Blyth.

[amazon_image id=”019982830X” link=”true” target=”_blank” size=”medium” ]Austerity: The History of a Dangerous Idea[/amazon_image]

Actually, the first few chapters start with the ‘dangerous idea’ part, with the author’s arguments about why austerity (ie. cutting the government’s budget deficits to reduce the level of its debt) is a bad thing in general, and a particularly bad thing when everyone tries to do so at the same time. This part will be somewhat familiar to readers of Paul Krugman’s blog, or Jonathan Portes on this side of the Atlantic. It overlooks some points I think are important – for example, glossing over the way tax increases and spending cuts will have different distributional implications; or ignoring the effects of inflation on real wages for low earners to focus on the redistribution from savers to borrowers. I also don’t agree with his argument about the specific causes of the financial crisis, which he pins on the securitised mortgages and the US repo market, but that’s not the heart of the book. Besides, Blyth is surely right to say morality tales about lazy Greeks and virtuous Germans, and other similar tropes of public debate about the crisis, do not amount to an economic analysis.

The main section is far more interesting, an account of the history of the idea that austerity is a good policy, that reducing the debt burden only requires reduced borrowing and less government. He traces the idea back to the late 17th century and ranges over the continent as well as the US and UK. Although there’s no mistaking this author’s political perspective, there is plenty of interesting material in this section. The book ends by asking whether or not current austerity policies will work. When the IMF has now said not, and a great majority of economists advocate bringing forward necessary infrastructure investment, Blyth’s answer will come as no surprise, albeit expressed more colourfully: “The deployment of austerity as an economic policy has been as effective in bringing us peace, prosperity and crucially a sustained reduction of debt as the Mongol Horde has been in furthering the development of dressage.”

The book will give opponents of the austerity strategy more ammunition, if they want it. I’m not sure it will change the mind of any proponents of the policy, however, given how obvious the conclusion is from page one. But then, as Blyth argues, this is not a rational economic debate. Austerity has a different kind of hold on its advocates.