Recessions and recoveries: a historical perspective

[Note: this is now slightly out of date. An updated version is here.]
Following the GDP numbers published January 25, here is a further updated version of NIESR's chart showing the path of recession and recovery in various previous downturns.  The chart shows that this "depression" - defined, admittedly somewhat arbitrarily, as the time period during which output remains below its previous peak, shown as the X-axis above - is now longer than that experienced during the Great Depression, and is not likely to end any time soon. It also shows how what was initially a reasonably strong, albeit patchy, recovery stalled in the autumn of 2010; since then there has been very little growth .  


Lonely Joe Parker's picture

It's interesting that the initial gradient from the start to ca.3mths is roughly the same in all cases. Is that due to something analagous to a degrees-of-freedom / measurement effect, or a deeper reason to do with how economies collapse, I wonder?

obelix's picture

This is completely incorrect, for the reason that you note. The "great depression" in the UK started in 1919 and ended in 1939. The economy fluctuated between various degrees of terrible during that 20-or-so year period. The current UK recession will not start looking like the UK great depression unless it lasts until 2018.

Unknown's picture

I had not realized that the U.K. Depression started in 1919. I've been working under the assumption that we were Britain circa 1930 and China was America 1930. Now it all makes sense that we are actually Britain circa 1926 and China is America 1926. This actually makes a lot more sense to me. So roughly, in 2028 we'll be in okay shape as long as some crazy person doesn't decide we are the enemy that must be destroyed and drags us into all consuming war where our very existance is threatened. Cool.

payguy's picture

Don't fall for the idea that tax revenue are required to pay-off government debt. In fact, it is a myth that taxes "pay" for any government spending.

When an economy is at full capacity (i.e. very low unemployment and all resources in the economy being used productively), a government may wish to spend say £20Bn on something everyone agrees is needed -it could be repaying govt debt, defending the country, building hospitals, whatever. When it spends this money it inevitably causes inflation - this is because you have more spending chasing the same amount of goods and services. The amount of goods and services does not change because the economy is already at full capacity.

To enable the government to spend without causing an inflationary spiral, the government taxes by an equal amount to prevent the private sector spending by the same amount -so overall the spending (public and private) remains roughly constant, so no inflationary spiral.

So the extra tax is to prevent an inflationary spiral when the economy is at full capacity - it is not required to "finance" govt spending. This is why govt economics is nothing like household economics.

However, when an economy is the position ours is in with excess capacity, spending by government is permissible without taxation as it doesn't cause inflation.

Given that our economy has not been at full capacity for over 30 years (hence the high unemployment), the government does not need to increase taxes or cut spending elsewhere to "pay" the interest on govt debt or to "pay" for anything.

The big question is why does the government issue bonds at all and pay interest to private investors? Why doesn't the govt just create the money at the mint or Bank of England - this won't be inflationary as there is spare capacity.

An answer often given is that when governments issue bonds someone has to surrender money to the government. If it wasn't for the bond that money would probably have gone into the banking system instead. This is called a 'reserve drain' and was clearly necessary when we had the Gold Standard/Bretton Woods or some other type of Fixed Exchange Mechanism.

The argument given now is that debt is a better way to stimulate the economy. Supposedly there is a problem with a liquidity trap in the banking system. By issuing bonds the government can take money away from the banking system and make sure that it is being spent.

However, it's pretty obvious that for countries with their own floating currency, deleveraging banks and with economies working at way, way below spare capacity that you can use QE to clear government debt at will without any inflationary effects.

This is obviously so in the UK since there is £275 billion sitting in the Asset Purchase Facility. This money was bought using reserve crediting in 2010/11 and the result of the purchases was deflationary - M4 last year after £200 billion of QE had hit stall speed with growth at only 2% (more than 5% growth is needed to prevent the economy contracting).

So the Tories are moaning about the huge and "unaffordable" government credit card bill. At the same time over a third of the debt they are moaning about is stuck in the government owned APF with no hope of it ever being anything other than cancelled and retired. To add to the hilarity the DMO pays interest on the £200 billion in the APF to the wholly government owned APF. This money is just building up and will eventually (as all profits for the Bank are) be returned to the taxpayer. You couldn't make this up.

So clearly in economic circumstances such as now you can print money directly, buy outstanding government debt and retire it with no inflationary consequences. Nevertheless Govts are continuing to use an explanation built up at a time of Bretton Woods to explain why they don't use the QE to clear down debts.

bsanchez's picture

Lucky Brits. Just 4 years in the 1930s to get out of the hole. Compare it with 26 years for Spain.

obelix's picture

Krugman has also been posting graphs and arguing that the great recession has now lasted longer than the Great Depression in the UK. With the risk of repeating my earlier comment, I think that's simply not true.

Britain was in its Great Depression the entire inter-war period from 1919-1939. The wars were terrible times as well, but it's hard to do good national accounting during wartime. It's safe to say that from a consumer's perspective the Great Depression at least the period 1914-1945. The real output per adult did not grow at all from 1919 to 1939, let's repeat that, zero real growth for two decades. The period of Great Depression in the US was merely a blip in the bottom of a true, multi-decade UK Great Depression.

The reasons for this multi-decade depression were many: Shrinking cash flow from the empire with increasing enforcement costs, overvalued gold standard exchange rate that required contractionary monetary policy, counterproductive housing subsidies that reduced the incentives to move from depressed regions to growing regions, generous increases to unemployment subsidies, decision to pay American's back about 90 cents on the dollar of WWI debts, etc. Unemployment benefit to average wage ratio, in particular, went from about 20% in 1919 to about 55% in 1939, but there were many other reasons as well.

Saying that the UK great depression was shorter than the current great recession displays some combination of ignorance and dishonesty.

Caralladas's picture

The Spanish crisis has gone further than the Great Depression, too. However, the 20th century witnessed a much worst crisis, at least for Spain...

payguy's picture

Joanathan- you should read some MMT. The basic argument would go something like-

There are only two ways to create money in the UK economy:-

1) The normal process of credit creation carried out by banks - banks lending out more money that they charge interest on. Apart from minting coins or printing notes this usually creates over 95% of the money (called M4) in the economy.

2) QE - The Bank of England crediting its reserves with money and then using the credits to buy assets or outstanding government debt from banks.

Since 2008 banks have largely shut down credit creation. M4 which normally grows at over 5% per year is only growing at 2% per year without QE. Expansion of below 5% pa means the economy contracts. This is by and large, as in all recessions after a financial crash is why the World and UK economies are in such a mess - deleveraging of the financial sector.

The line that our Government (and now several others) are giving us is that There Is No Alternative to austerity and cuts. They are justifying massive tax rises and catastrophic cuts in public spending because they say excess government debt, built up due to the massive worldwide recession in 2008 and the cost of bank bail outs, must be paid down.

This is obviously false as Governments can use QE to buy up Government debt from the banks that are holding it and retire it. This is happened to a massive degree already in the UK with over a third (over £275 billion) of the UK's Government debt is currently sitting in the wholly publicly owned Asset Purchase Facility.

But what about inflation? Wont retiring Government debt in this way cause inflation? No- if there was inflation it would happen when the Bank of England bought the Government debt up from the banks. This is the moment reserve credits are released and there is an increase in bank liquidity. We have done £275 billion of QE (equivalent to about 20% of UK GDP) since 2009 and M4 has contracted and we are at risk of deflation rather than inflation.

Quite simply no matter what we say to them banks don't want to lend enough to get the economy growing.

So it is perfectly safe to retire Government debt when banks aren't creating enough credit in the economy. If this is a natural phenomena because the banks don't want to lend (they are deleveraging) it is safe to retire Government debt. As long as the money supply is kept at around 5% all is well - the economy neither contracts too quickly causing inflation or collapses causing a depression.

It is also perfectly safe at a happier time in the economic cycle. Say in 5 years time when the economy is expanding, banks are lending too much. At this point we would want to use QE to expand the money supply as we would want to restrict bank lending to ensure we never get another crunch like 2008. When eventually we need to increase the capital adequacy levels in banks (to make them safe) we will need some mechanism to ensure the money supply is kept expanding at the needed rate (5%) we will need to use QE to retire government debt.

The argument then is about productivity and crowding out. But that is by no means an argument you can make when you are reducing the size of the state now and causing a depression through unnecessary and destructive fiscal contraction in the face of a liquidity trap.

The most depressing thing about this is that the current Government is misleading people so badly about how the money supply and economy work. The analogy of a National economy and household one is - the only word I can think of is evil. This is an action of people who represent their corporate funders and want to mislead you to prevent you questioning in who's interest they are acting.

thebicyclethief's picture

just wanted to say thanks to all comments for a very interesting, balanced and informative discussion.

Jonathan Harris's picture

Jonathan's initial blog article, graphically highlighting the extent of the current recession, compared to the most recent previous ones in history is linked to the second set of comments by Payguy, which is seeking to both highlight some of the causes of the protracted nature of the current extension of the recession's effects and posits a hypothesis about potential policy errors contributing to it.

Is there any data available for the UK recession of the early 1920s, which might also be included in future comparisons of this nature? - I realise that this was not a global recession. It may provide answers or help clarify the points raised by obelix above about the timing of recessions and Depressions from 1919-1933.

In view of the high risks of continuing UK economic policy error, is there no way that NIESR, and the Independent Bank of England, could highlight these points more clearly, with more data and substantive analysis and alternative policy prescription in the public domain?

I guess additional credibility might come from university academic economists through supporting articles in OER and or REJ.

obelix's picture

I have data from Chapman and Knight (1953) for the UK. It starts in 1920 and ends in 1938, roughly the same period. The exact dating by year is difficult, and data are noisy so I wouldn't make too big of a deal about the exact starting point of the depression. The whole table is hard to reproduce, so I am just going to 1920, 1921, 1922, 1929, 1933, and 1938 data points because one can kind of linearly interpolate between those and not go too far off.

Unemployment percentage:

Average annual earnings (index 1939 = 100)
1920: 132.5
1933: 96.4
1938: 105.7

Retail prices (index 1939 = 100)
1920: 157.6
1921: 143.0
1922: 115.8
1929: 103.8
1933: 88.6
1938: 98.7

By these measures, the 1929-1933 period wasn't really even a depression in the UK by the Wikipedia -10% definition if considered in isolation. If considered relative to the 1920 values, however, the 1929-1933 period was very much part of the depression that started in 1920.

Here's some much more scrubbed data from a modern source. The data are normalized per adult instead of per capita, which more accurately reflects business conditions. The data are from a Fed paper ( and I am eyeballing the data from the Figure 1:
Output per adult in UK (index 1905 = 100)
1919: 113
1920: 101
1921: 94
1922: 87
1929: 99
1933: 93
1938: 110

Again, it's safe to interpolate between the listed points linearly. As before, the story is that there was a depression, which consisted of decline 1919-1922 (from 113 to 87) and a bumpy recovery 1922-1938 (from 87 to 110) that almost recovered the per adult real output.

To summarize, the period coinciding the US Great Depression doesn't qualify as a depression in the UK by the Wikipedia rule. By any sensible definition that period in the UK is part of the long recovery stage of the depression that started in 1919.

Jonathan Harris's picture

Obelix - You have omitted the data between the periods of the two recessions above. Whilst output may not have grown to exceed the levels prior to the 1919-22 recession during the 20's, it did not continue to decline as per that recession and the Great Depression.

A better description of the economy between 1922 and 1929 would be one of stagnation, rather than depression or recession. Perhaps if you shared the data with us, it would illustrate that?

Sarah C.Rodriguez's picture

Wow! Thank you! I permanently wanted to write on my blog something like that. Can I take a part of your post to my blog?
Yo-Zuri Custom Inshore Popping Pack

Katina E.Coulombe's picture

I'll right away snatch your rss feed as I can not find your email subscription link or newsletter service. Do you have any? Please allow me understand in order that I may just subscribe. Thanks.
La Canadienne Women's Felicia Boot

obelix's picture

I omitted some of the the Chapman and Knight data because I had to type them into the post. I picked minor inflection points such that it's safe to linearly interpolate between the points I gave. Scouts honor, you are not missing much if you just use a ruler there. For extra assurances, take a look at the Figure 1 of this source for example:

The US Great Depression period in the UK was a period of bumpy recovery from the collapse after the World War. Based on output per adult, there was only one year of decline in the UK during the US Great Depression. That period, in isolation, is a normal recession not a depression. In the broader context, it's part of a slow and bumpy recovery.

talha's picture

USA economic indicators suggest that the economic recovery, may be picking up speed. The national unemployment rate in February held steady at 8.3 percent. For the third consecutive month the economy gained more than 200,000 jobs in February; the last time the nation saw three straight months of jobs gains of this magnitude was in 2006. During the last quarter of 2011, inflation-adjusted GDP grew at an annual rate of 3 percent, well above its growth rates for the previous three quarters of the year.

plastic card holders
plastic business cards
scratch cards printing

Muhammad Atif's picture

Saying that the UK great depression was shorter than the current great recession displays some combination of ignorance and dishonesty.
Plastic card
Plastic card
Hank Hendricks

Total views: 34,081