It’s not all doom and gloom. Beware politicised forecasts
9 February 2017
SUGGESTED
Economic Theory
Uncategorized
10 February 2017
Tax and Fiscal Policy
19 November 2024
I was on The Big Questions (BBC1) recently talking about poverty. Specifically I was trying to help answer the question ‘Will more children be raised in poverty?’ The topic may have been inspired by a report from the Resolution Foundation released this week which predicted falling living standards and rising inequality for the rest of the decade.
The Resolution Foundation report echoed the pessimism of a 2016 report from the Institute of Fiscal Studies which predicted no income growth for people in lower income brackets and a rise in absolute child poverty by 2020 (‘absolute poverty’ is defined as having an income that is 60% or less of the 2010/11 average, after adjusting for inflation). Both reports claimed that these negative outcomes could be avoided if the government abandons its plans to cut some out-of-work benefits. The IFS has since made headlines by claiming that it will take until 2021 for wages to return to 2008 levels thanks to Brexit.
I am sceptical about these gloomy forecasts. The Resolution Foundation and the IFS have been saying the same thing for so long that one day they are bound to be right, but that is the problem. They said the same thing when the government cut benefits at the start of the era of supposed ‘austerity’. Since 2011, if not earlier, they have been predicting a wave of inequality, poverty and falling incomes. It’s 2017 and it still hasn’t happened.
If somebody kept offering me useless horse-racing tips, I would stop listening to them. Given their conspicuous failure to read the tea leaves of ‘austerity’ correctly, it is surprising that the media continue to take the IFS’s prognostications so seriously. I suppose there is always a market for doom and gloom, especially when you are ‘respected’ and ‘independent’ – as the media insist the IFS is.
Bad predictions are easily forgotten so let us recap. In 2011, the Institute for Fiscal Studies predicted that by 2014/15, absolute child poverty would have risen from 17.5% to 24% (similar predictions were made in this 2011 IFS report and this 2012 report for the Joseph Rowntree Foundation). In fact, it fell to 17%.
Actually it was worse than that. IFS modelled the data based on six different scenarios – high employment/low employment, benefit cuts/no benefit cuts, progressive/regressive earnings growth – but all but one of them predicted that the rate would be at least 23% by 2015/16. The sole exception was the ‘no tax and benefit reforms’ scenario but even that predicted a rise to 18.5%. In other words, they had six chances to get it right and overshot every time.
The IFS also made predictions about what would happen to incomes between 2007/08 and 2015/16. The outlook was grim, with every income group worse off by 2015/16 under almost every scenario. The sole exception was a small number of people at the bottom end whose incomes would rise slightly if the government abandoned its welfare reforms (are you starting to see a pattern here?).
In 2011, the IFS argued that people in the middle and high income brackets had already seen their incomes fall whereas the worst was yet to come for lower earnings. For ‘those towards the bottom,’ they said, who were ‘dependent more on benefit incomes, falls in income will happen largely between 2010–11 and 2015–16’.
The Resolution Foundation struck a similar note in a 2011 report which predicted that incomes would remain below pre-crash levels for the rest of the decade and that incomes would fall to twentieth century levels for many poorer groups.
This was followed in 2012 with the Resolution Foundation’s Gaining From Growth report which predicted lower incomes for everybody except the rich. ‘Even on optimistic growth assumptions,’ they said, ‘low income households in 2020 now look likely to have incomes 15 per cent below those in 2008’. They predicted that middle income households would have incomes three per cent below 2008 levels while richer groups would see a rise.
It is too early to categorically debunk the Resolution Foundation’s forecast for 2020, but Office for National Statistics data allow us to put the IFS’s predictions to bed. There was income growth across the distribution between 2007/08 and 2015/16 with the exception of the richest fifth who saw a 3% decline. The poorest fifth saw the biggest rise: 13 per cent above pre-crash levels by 2015/16.
As the ONS notes, median incomes are already £1,000 higher than they were before the crash:
“After taking account of inflation and changes in household structures over time, the median disposable income has increased by £600 (or 2.2%) since 2014/15 and is £1,000 higher than the pre-economic downturn level observed in 2007/08.”
It seems to me that Office for National Statistics data should be more prominent in public debate than speculation from think tanks, but that is not how it works. I suspect that many people would be surprised by the evidence shown by the ONS. Similarly, many people would be surprised to hear that rates of poverty and inequality have fallen in recent years. These are verifiable facts and yet it was the Resolution Foundation’s hunch that inequality might finally rise that made the front pages this week.
In all fairness, economic forecasting is not easy. For a start, you have to rely on forecasts from other organisations about such variables as GDP, unemployment and inflation – and these are often wrong too. But the IFS/Resolution forecasts have not just been wrong, they have been woefully and utterly wrong. We were told inequality would rise. It fell. We were told poverty rates would rise. They fell. We were told incomes would fall. They rose. We were told the rich would do better than the poor. They did worse.
This is a less effective system than flipping a coin. It suggests something more than bad luck. It suggests a systemic bias towards doom and gloom.
A forecast is only as good as the assumptions it is based on and one of the flawed assumptions of the IFS and Resolution Foundation reports is the belief that benefit cuts will necessarily result in lower incomes for those who are on welfare. But this does not account for people changing their behaviour as a result of incentives. One of the reasons income growth has been particularly strong among the poorest fifth of households is, as the ONS has explained, that people have come off benefits and gone into work – and work pays more than benefits.
Now the Resolution Foundation and IFS are back again, without a word of contrition, to double down on their forecasts for 2020. This time, they tell us that living standards/poverty/inequality are going to get worse because Brexit-inspired inflation will wipe out the effect of wage increases. Anything is possible, of course, but it is equally possible that employers will take inflation into account when offering pay rises. Businesses do this when inflation is in double digits so it does not seem unthinkable that they will to do it if inflation rises to the dizzy heights of two or three per cent (it is currently 1.6 per cent).
I am aware that I am leaving a hostage to fortune by writing this post. The economy could go belly up at any minute due to some unforeseeable shock and I’ll be left with egg on my face, but we are talking here about probabilities. I believe it is probable that real incomes will be higher in 2020 than they are today and that absolute poverty will be lower. I make no prediction about the nonsense of relative poverty and I don’t care about inequality, but if the economy continues to grow, it would be surprising if the IFS and Resolution Foundation are correct in their predictions.
Real incomes track GDP very closely and the historical record shows that growth is more likely than recession in any given year. The rate of inflation or the state of the welfare system have a trivial effect on real incomes in comparison to GDP growth. A rising tide lifts all boats. If you had picked any year at random from the last two centuries and were forced to bet on whether incomes would rise or fall in the next decade, you would be a fool to bet on a fall.
A longer version of this article was posted on the author’s blog ‘Velvet glove, iron fist’.
The Resolution Foundation report echoed the pessimism of a 2016 report from the Institute of Fiscal Studies which predicted no income growth for people in lower income brackets and a rise in absolute child poverty by 2020 (‘absolute poverty’ is defined as having an income that is 60% or less of the 2010/11 average, after adjusting for inflation). Both reports claimed that these negative outcomes could be avoided if the government abandons its plans to cut some out-of-work benefits. The IFS has since made headlines by claiming that it will take until 2021 for wages to return to 2008 levels thanks to Brexit.
I am sceptical about these gloomy forecasts. The Resolution Foundation and the IFS have been saying the same thing for so long that one day they are bound to be right, but that is the problem. They said the same thing when the government cut benefits at the start of the era of supposed ‘austerity’. Since 2011, if not earlier, they have been predicting a wave of inequality, poverty and falling incomes. It’s 2017 and it still hasn’t happened.
If somebody kept offering me useless horse-racing tips, I would stop listening to them. Given their conspicuous failure to read the tea leaves of ‘austerity’ correctly, it is surprising that the media continue to take the IFS’s prognostications so seriously. I suppose there is always a market for doom and gloom, especially when you are ‘respected’ and ‘independent’ – as the media insist the IFS is.
Bad predictions are easily forgotten so let us recap. In 2011, the Institute for Fiscal Studies predicted that by 2014/15, absolute child poverty would have risen from 17.5% to 24% (similar predictions were made in this 2011 IFS report and this 2012 report for the Joseph Rowntree Foundation). In fact, it fell to 17%.
Actually it was worse than that. IFS modelled the data based on six different scenarios – high employment/low employment, benefit cuts/no benefit cuts, progressive/regressive earnings growth – but all but one of them predicted that the rate would be at least 23% by 2015/16. The sole exception was the ‘no tax and benefit reforms’ scenario but even that predicted a rise to 18.5%. In other words, they had six chances to get it right and overshot every time.
The IFS also made predictions about what would happen to incomes between 2007/08 and 2015/16. The outlook was grim, with every income group worse off by 2015/16 under almost every scenario. The sole exception was a small number of people at the bottom end whose incomes would rise slightly if the government abandoned its welfare reforms (are you starting to see a pattern here?).
In 2011, the IFS argued that people in the middle and high income brackets had already seen their incomes fall whereas the worst was yet to come for lower earnings. For ‘those towards the bottom,’ they said, who were ‘dependent more on benefit incomes, falls in income will happen largely between 2010–11 and 2015–16’.
The Resolution Foundation struck a similar note in a 2011 report which predicted that incomes would remain below pre-crash levels for the rest of the decade and that incomes would fall to twentieth century levels for many poorer groups.
This was followed in 2012 with the Resolution Foundation’s Gaining From Growth report which predicted lower incomes for everybody except the rich. ‘Even on optimistic growth assumptions,’ they said, ‘low income households in 2020 now look likely to have incomes 15 per cent below those in 2008’. They predicted that middle income households would have incomes three per cent below 2008 levels while richer groups would see a rise.
It is too early to categorically debunk the Resolution Foundation’s forecast for 2020, but Office for National Statistics data allow us to put the IFS’s predictions to bed. There was income growth across the distribution between 2007/08 and 2015/16 with the exception of the richest fifth who saw a 3% decline. The poorest fifth saw the biggest rise: 13 per cent above pre-crash levels by 2015/16.
As the ONS notes, median incomes are already £1,000 higher than they were before the crash:
“After taking account of inflation and changes in household structures over time, the median disposable income has increased by £600 (or 2.2%) since 2014/15 and is £1,000 higher than the pre-economic downturn level observed in 2007/08.”
It seems to me that Office for National Statistics data should be more prominent in public debate than speculation from think tanks, but that is not how it works. I suspect that many people would be surprised by the evidence shown by the ONS. Similarly, many people would be surprised to hear that rates of poverty and inequality have fallen in recent years. These are verifiable facts and yet it was the Resolution Foundation’s hunch that inequality might finally rise that made the front pages this week.
In all fairness, economic forecasting is not easy. For a start, you have to rely on forecasts from other organisations about such variables as GDP, unemployment and inflation – and these are often wrong too. But the IFS/Resolution forecasts have not just been wrong, they have been woefully and utterly wrong. We were told inequality would rise. It fell. We were told poverty rates would rise. They fell. We were told incomes would fall. They rose. We were told the rich would do better than the poor. They did worse.
This is a less effective system than flipping a coin. It suggests something more than bad luck. It suggests a systemic bias towards doom and gloom.
A forecast is only as good as the assumptions it is based on and one of the flawed assumptions of the IFS and Resolution Foundation reports is the belief that benefit cuts will necessarily result in lower incomes for those who are on welfare. But this does not account for people changing their behaviour as a result of incentives. One of the reasons income growth has been particularly strong among the poorest fifth of households is, as the ONS has explained, that people have come off benefits and gone into work – and work pays more than benefits.
Now the Resolution Foundation and IFS are back again, without a word of contrition, to double down on their forecasts for 2020. This time, they tell us that living standards/poverty/inequality are going to get worse because Brexit-inspired inflation will wipe out the effect of wage increases. Anything is possible, of course, but it is equally possible that employers will take inflation into account when offering pay rises. Businesses do this when inflation is in double digits so it does not seem unthinkable that they will to do it if inflation rises to the dizzy heights of two or three per cent (it is currently 1.6 per cent).
I am aware that I am leaving a hostage to fortune by writing this post. The economy could go belly up at any minute due to some unforeseeable shock and I’ll be left with egg on my face, but we are talking here about probabilities. I believe it is probable that real incomes will be higher in 2020 than they are today and that absolute poverty will be lower. I make no prediction about the nonsense of relative poverty and I don’t care about inequality, but if the economy continues to grow, it would be surprising if the IFS and Resolution Foundation are correct in their predictions.
Real incomes track GDP very closely and the historical record shows that growth is more likely than recession in any given year. The rate of inflation or the state of the welfare system have a trivial effect on real incomes in comparison to GDP growth. A rising tide lifts all boats. If you had picked any year at random from the last two centuries and were forced to bet on whether incomes would rise or fall in the next decade, you would be a fool to bet on a fall.
A longer version of this article was posted on the author’s blog ‘Velvet glove, iron fist’.
SHARE
