Economic history · The 2008 Financial Crisis
Seven steps, about ten minutes, and a question at each. You need no economics and no preparation, only a willingness to guess before you are told.
Britain announced on 8 October 2008 that it would put public money into its banks as shares rather than loans, and the whole of the support peaked at over a trillion pounds.
The banks were rescued and then lending shrank anyway. British output fell 6 per cent and was slow to come back, and yet unemployment here stopped well short of America's.
Seven steps on where the difference went, and on the one thing that never returned.
What this module covers
Written for every level. Tap any underlined word for what it means, and open the boxes below for the economics behind each decision. If you already know the theory, skip both and the history reads straight through.
Step 1 of 7
Bank lending to British households and companies grew by 3.8 per cent across 2008, averaged over the year, while the rescues were going on.
Money out on loan is a stock, like water in a tank: new lending runs in and what people pay back runs out. The stock shrinks only when they pay back faster than anyone borrows.
The Bank of England's own trend for that stock is growth of 5.8 per cent a year.
Take a guess. You are not expected to know the answers, and being wrong is what makes a number stick.
Lending grew 3.8 per cent across 2008. What did it do in 2009?
The Bank of England put a figure on how much of the 2009 fall was banks choosing to lend less rather than customers asking for less.
Both would show the same thing: less money out on loan. What tells them apart is the price. If customers had simply gone quiet, banks chasing the ones left would have cut what they charged. Instead the extra they charged over Bank Rate went up and stayed up, which is what a lender does when it wants to lend less rather than more.
A percentage point is a step on a percentage scale: growth of 5 per cent falling to 3 per cent is a drop of two points. The Bank counts what moved lending growth in points of that kind.
Banks price a loan as an extra on top of Bank Rate, the the Bank of England sets, and how big that extra is says which of the two was happening.
The Bank's trend for lending growth is the same 5.8 a year that step one gave, and lending in 2009 came in 0.7 per cent below where it started.
How many percentage points did the Bank of England reckon came off lending growth in 2009 because the banks were lending less?
For scale. The same estimate for 2008, the year the banks were being rescued, is 1.4 percentage points.
The American that followed the crisis ran from December 2007 to June 2009, on the dating of a committee at the National Bureau of Economic Research.
The committee waits until the figures have settled before it decides, so a recession is normally declared over long after it is.
Before this one, the longest American recessions since the second world war had each run sixteen months, in 1973 to 1975 and in 1981 to 1982.
Dating one takes time: this recession ended in June 2009 and was not declared over until September 2010, fifteen months later.
How many months did the American recession of 2007 to 2009 run?
For scale. The two longest before it had run sixteen months each.
British peaked at the start of 2008 and reached its bottom in the middle of 2009, about 6.45 per cent lower on the current figures.
Getting back means the country producing as much in a quarter as it did before, which is a lower bar than the one it would have reached if nothing had happened.
The national accounts are revised for years afterwards, and they have put that fall at 6.3 per cent and at 6.45 per cent, and the climb back at twenty-one quarters and at twenty-two.
Getting back to 2008 is not the same as making up the growth of the five years in between, which is lost whatever the accounts eventually settle on.
How long was it before British output was back where it had been at the start of 2008?
British was 5.2 per cent at the end of 2007, before any of it started.
The unemployment rate counts people out of work and looking for it, as a share of those working or looking. It misses somebody who has given up, and somebody on short hours who would rather work more.
American unemployment reached 10.0 per cent in October 2009, on the figures as they were later revised.
British unemployment was 5.2 per cent before the recession. What did it reach by the start of 2010?
The Bank of England set this recession against the two before it. Output fell 6.2 per cent this time against 2.5 per cent in the recession of the early 1990s.
A firm facing a bad year can cut its wage bill by paying fewer people, by paying people for fewer hours, or by paying them less an hour. The three add up to the same saving and land on different people.
Employment fell 1.9 per cent in this recession, and the hours an average worker put in fell 2.2 per cent on top of that. Hourly pay went up 0.1 per cent, which is to say it stood still.
Employment fell 1.9 per cent in this recession. In the recession of the 1990s, which took less than half as much off output, was the fall in employment higher or lower?
A tie counts as correct either way.
For scale. Employment fell 2.4 per cent through the recession of the early 1980s, when output fell 4.7 per cent.
British , which is what a worker produces in an hour, had grown steadily enough before 2008 that a line could be drawn through it and carried forward.
Keeping staff on through a bad year saves the jobs and costs the sums, because the same people each produce less while there is less to do. The figure does not climb back on its own when the work returns.
Six years on, output per hour was still around 4 per cent below the level it reached before the crisis, while output itself had broadly recovered.
The National Institute of Economic and Social Research finds “limited direct evidence linking banking impairment to productivity losses”, and puts the weakness inside businesses rather than between them.
By how much was output per hour below the line it had been on before the crisis?
For scale. Against its own pre-crisis peak, rather than against that line, output per hour was about 4 per cent short. The gap against the line is much the larger of the two.
Module 6 of 7 in The 2008 Financial Crisis
Lending to households and companies
up 3.8 per cent in 2008, down 0.7 per cent in 2009How much of that was the banks
8.3 percentage points off lending growth in 2009America's
18 months, the longest since the warBritish
down about 6.45 per cent, and slow to get backBritish
5.2 per cent to 8.0 per cent, against America's 10.0The trade
employment down 1.9 per cent here against 3.4 per cent in the 1990s, and hourly pay after inflation up 0.1 per cent against 7.3per hour
around 16 per cent below its old trendYou met six terms in this module
, , , , ,
Britain lost more of its output than America and fewer of its jobs, and what paid for that was a pay rise that never came.
[1] How much of the fall in lending was the banks lending less, on the Bank of England's own count: Bank of England, Understanding the weakness of bank lending, Quarterly Bulletin 2010 Q4. Read twice off the PDF, identical both times. The decomposition, in percentage points, averages for 2008, 2009 and 2010 H1: "Credit supply shocks -1.4 (-3 to 4) -8.3 (-9 to -4) -6.1 (-7 to -2)"; "Other shocks -0.6 (-5 to 1) 1.8 (-2 to 2) -2.6 (-6 to -2)"; "Trend 5.8 (5 to 6) 5.8 (5 to 6) 5.8 (5 to 6)"; "Lending(b)(per cent) 3.8 -0.7 -3.0". And on prices, differences from 2004-07 averages: "Mortgage spread(c) 0.8 2.7 2.5". THE ARITHMETIC CLOSES: 5.8 - 8.3 + 1.8 = -0.7. The article is careful that this is one estimate with a wide range and that demand weakened too, and the module says both.
[2] The argument on the other side: International Monetary Fund, United Kingdom: 2010 Article IV Consultation, Staff Report, 21 October 2010. NOT RE-FETCHED THIS SESSION, and that is why it has its own id rather than being hung on the Bank's. The quotation is the spine's, from its signed source note: weak lending "reflected not only tight supply, but - perhaps more importantly - weak demand for investment finance". It is the counter-argument the treatment says must be carried, and the module carries it in the reveal of the step that gives the Bank's figure. A later session should open the Staff Report directly.
[3] How long the American recession ran, and how that compares: National Bureau of Economic Research, Business Cycle Dating Committee announcement, 20 September 2010. Read twice, identical both times: "The trough marks the end of the recession that began in December 2007 and the beginning of an expansion. The recession lasted 18 months, which makes it the longest of any recession since World War II. Previously the longest postwar recessions were those of 1973-75 and 1981-82, both of which lasted 16 months." The 16 months is what the reader is given to reason from under RULE 64.
[4] How high American unemployment went: US Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, series LNS14000000. NOT RE-FETCHED THIS SESSION; the figure is the spine's, from its signed source note. 10.0% in October 2009 on the revised series. THE FIRST RELEASE FOR THE SAME MONTH, on 10 November 2009, SAID 10.2%, and the two must never appear together without their vintages. Only the revised 10.0% prints.
[5] How far British output fell and how long it took to get back: Office for National Statistics, series ABMI, Gross Domestic Product: chained volume measure. Read twice off the series' own data endpoint, identical both times. On this vintage the peak is 2008 Q1 and the trough 2009 Q2, a fall of 6.45%, and output first passes the old peak in 2013 Q3 - twenty-two quarters. THE SAME INSTITUTION HAS PUBLISHED -6.3% AND 21 QUARTERS (Communicating the UK Economic Cycle, 11 November 2022) AND 22 QUARTERS (2018). Three vintages of one number, which is why step 4 asks in years: five and a bit is true on all of them, and no vintage of the accounts makes it four or six.
[6] How high British unemployment went: Office for National Statistics, series MGSX, unemployment rate, aged 16 and over, seasonally adjusted. Read twice off the series' own data endpoint, identical both times: 5.2% in 2007 Q4, 6.4% in 2008 Q4, 7.8% in 2009 Q4, 8.0% in 2010 Q1 and 8.4% in 2011 Q4. The 8.0% is the peak inside the recession and its aftermath; the 8.4% two years later is the highest the series reaches.
[7] The trade, measured against the two recessions before it: Bank of England, Changes in output, employment and wages during recessions in the United Kingdom, Quarterly Bulletin 2010 Q1. Read twice off the PDF, identical both times. Table A, "Output, employment, hours and wages during recessions", per cent: "1980s 1990s 2000s GDP(c) -4.7 -2.5 -6.2 ... Employment(d) -2.4 -3.4 -1.9 Average hours(e) -3.0 -1.9 -2.2 ... Real hourly wages(f) 2.7 7.3 0.1". The article's own prose says "UK output is estimated to have fallen by 6%", which is the table's -6.2 rounded, and the module prints the table. NOTE THE -6.2 AGAINST THE ONS'S -6.3: different vintages and different peak quarters, both printed, each with its own source.
[8] What never came back: Bank of England, The UK productivity puzzle, Quarterly Bulletin 2014 Q2. Read twice off the PDF, identical both times: "whole-economy output per hour remains around 16% below the level implied by its pre-crisis trend", and "Even six years after the initial downturn, the level of productivity lies around 4% below its pre-crisis peak, in contrast to the level of output, which has broadly recovered to its pre-crisis level." The article also attributes 3 to 4 percentage points of the shortfall to reduced investment in physical and intangible capital, which does not print: it is a share of the PRODUCTIVITY shortfall rather than of the output fall, so F1 does not forbid it. It prints in step 7's reveal; an earlier draft of this note said it did not, and a compliance reader refused the note for contradicting the build.
[9] The argument that the banks are not the explanation: National Institute of Economic and Social Research, on the productivity puzzle, May 2014. NOT RE-FETCHED THIS SESSION; the quotation is the spine's, from its signed source note: NIESR finds "limited direct evidence linking banking impairment to productivity losses" and that "the main driver of the weakness of productivity is the within-business component". Carried because the treatment says the disagreement at link 40 is carried, and because without it step 7 would read as though the banking crisis explained the productivity shortfall, which no institution claims.
Country data from the World Bank (CC BY 4.0) and the UNDP Human Development Report (CC BY 3.0 IGO)
Games · Learn · Atlas · Privacy