The rise in the number of bureaucrats while the number of people at the "coal face" is exactly the phenomenon observed in Parkinson's Law.

There was a very intelligent sounding chap from the probation service on TV last night. He explained that there has been a trend in recent years away from longer sentences followed by release with no probation, to shorter sentences followed by periods of probation. He was quite frank that probation is a 'risk'. This may explain the rise in re-offending. Perhaps longer sentences followed by probation would be worth a try but, of course, this is resource heavy.

Posted by: HJHJ at December 6, 2006 09:56 AM

Does the evidence really support this theory?

If the welfare state is to blame for high levels of crime and imprisonment, then why is that the highest incarceration rates in the high income OECD countries are to be found in the USA (468.5 prisoners per 100,000 of population), New Zealand (132.3), Spain (110.7), Australia (93.4) and England and Wales (90.2) - coincidentally countries with relatively low levels of social spending as percentage of GDP? Likewise, why do the countries famed for their generous welfare states have much lower incarceration rates (e.g. Norway - 29.8, Netherlands - 34.9, Sweden - 41.3, Denmark - 42.9, Finland - 49.6)?

I'd be interested to hear your explanation for this...

Posted by: John Hudson at December 6, 2006 11:35 AM

I appreciate the point you make. My answer is this:

My claim is that the welfare state is a major cause of the vast increase in crime that has been experienced in Britain over the past century. There is too much evidence for this to put here (it is all in the book itself) but I will mention, for example, that the rise in crime has co-incided with or slightly lagged the creation of the welfare state (Churchill's legislation for national insurane for unemployment and so on). Those who agree that crime has risen in this time but disagree that the welfare state has been a major factor might like, I suggest, to offer a better, alternative explanation.

All the countries you mention have had welfare states and all the countries you mention, I strongly suspect without having all the figures to hand, have experienced major increases in crime rates over a similar time frame.

You suggest that Britain and certain other countries have low 'social spending' but high incarceration rates. You suggest that therefore there can be no serious causal relationship between welfare states and crime. If anything, the implication would be that welfare states reduce crime.

I do not know how you define 'social spending' but Britain, by any standard, has a major welfare state. It has state social security payments, state schooling, state housing, state healthcare and state pensions. I cannot agree that Britain is a minnow in the state welfare department.

It is not only social security which contributes to crime. State education and housing contribute, too. In these two cases it is fairly easy to provide evidence for a connection. It is more difficult to do so for healthcare and pensions but even these, I suggest, also have a small role because they contribute to the reduction in the need for people to show responsibility in the way they lead their lives.

A second important point is that the structure of social security matters a great deal. Social security is money given out on certain conditions. The amount of money matters and the conditions matter. Sweden, as I understand it, offers generous amounts of money to the unemployed but only on fairly tough conditions. After a certain period, an unemployed hospital consultant would have to take a job as a road-sweeper if need be. He would also have to move from a town with no jobs to another town with jobs.

Ireland, I believe, offers relatively generous support for unmarried mothers. Italy does not. This strongly affects behaviour in this regard in these two countries. So structure matters in social security. Taking a gross figure for 'social spending' is not good enough (and this is without mentioning the way in which 'social spending' figures can be misleading as with Mr Brown's 'tax credits' which he does not treat as 'social spending' although most people think are just that).

So I do not accept that the figures you cite invalidate my argument that our welfare state has been a major factor in the vast increase in crime in Britain in the past century.

Thankyou for your point which is certainly worth answering.

Posted by: James Bartholomew at December 6, 2006 02:03 PM

Thanks for the reply James.

I think it would be almost impossible to give a proper response to the historical issue you raise because data is too patchy outside of the modern era, but just as it would be necessary to take care over what we mean by social spending (as you rightly suggest), we also need to be careful in terms of what we mean by crime when looking at data too.

The social spending I was referring to was the percentage of GDP accounted for by public and social expenditure by the government, as defined by the OECD, which includes all social security and health programmes but not education. Actually, if we look at the OECD as a whole, the data do not suggest there is a clear and simple link (in either direction) between imprisonment rates and social spending, but the outliers at the top and bottom in terms of imprisonment tend to be lowest and highest social spenders respectively. I would say that something is going on, but would certainly accept that the issue is much more complex than social spending reduces imprisonment.

What is clear, however, is that lower levels of social spending do not reduce crime, but seem to be linked with comparatively higher levels of imprisonment. So, your argument does seem to be weakened in my view! Interestingly, while it could be true that, as you argue, the welfare state brought with it higher crime levels as it expanded (though I am not convinced), it might be equally true that its retrenchment or removal will not reverse those higher levels of crime in the present but, rather, increase them further. What worked in the past will not necessarily work in the present: we could both be right!

In terms of the details of the different national systems, I don't think your description of the Swedish system is fully accurate and, in truth, that the work focused measures for the unemployed in Sweden are pretty much like the UK's (and with similar patterns in terms of the duration of unemployment as a consequence), but with the major exception that unemployment benefits in Sweden are more generous. In Sweden, a single male worker who had been on average wages could expect to receive almost 80% of their previous income from unemployment insurance for 60 weeks and even after five years of unemployment could expect benefits to replace almost 70% of their previous income.

If we take a contrary example of a very tight unemployment support system - the USA - the same single male would receive generous support in the short term: around 60% of previous earnings would be replaced for six months. After this, typically only 7% of previous income would be replaced by benefits. For many low waged workers even the initially generous support would not exist because the qualifying conditions are pretty tight for unemployment insurance (previous earnings above a certain level, regular work record for instance). Consequently, for many low paid single males, losing your job means a sudden and drastic reduction in income will occur.

You argue that the structure of social security matters a great deal and I would wholeheartedly agree with you on this point. Do you think the differing structures of the unemployment systems in the USA and Sweden impact on the ten-fold variation in incarceration rates we see when we compare the USA and Sweden?

Posted by: John Hudson at December 6, 2006 07:26 PM

I'm glad that John has correctly noted that Sweden marries a very generous welfare system with very low incarceration rates.

...Which is no doubt why it also has the highest per capita crime rate in Western Europe.

Perhaps not everything is reducible to economics.
;)

Posted by: Paul H at December 6, 2006 11:36 PM

This is a fascinating and important area for debate. If, as I contend, the British welfare state has been a major factor causing the massive rise in crime in Britain, it matters. If, on the other hand, welfare states reduce crime instead, then that matters a lot, too.

The contrast John makes between the USA and Sweden is interesting and challenging. Unfortunately I suffer a major handicap in comparing the two countries. I am not expert in either of them. Yes, I know more about their social security systems and history than the man in the street. But I do not feel I have reached a level of knowledge and understanding that would enable me to be dogmatic in any comparison. It has been a big job to achieve what knowledge I have reached about the history and current state of Britain and its welfare state. The welfare state comprises many areas and the history in each area is long. Establishing links and causes that cut across these areas is not easy. I believe they are there but they tend to be hard to prove. I have not had the time to gain a similar knowledge of all the other countries in the OECD!

I don't seek to avoid John's point. Only to confess that I wish I was better equipped to give it as good an answer as it deserves.

I would make the brief observation that, of course, a few statistics can be misleading, but I won't rely on that for my answer.

I suppose at the core of my response is, "Of course, welfare states are not the cause of all crime. Of course, there are other factors, too."

I have recently been reading how dangerous life was in Renaissance Italy. Think how perillous it must be to live now in certain African countries. Law has broken down. In both these cases, 'social spending' was or is minimal. You could say in each case that the welfare state was or is tiny and yet there was or is a great deal of crime. But, I suggest, that would not prove that welfare states do not promote crime. It would only show that other factors can promote crime, too.

The USA and America have markedly different histories, to put it mildly. For most of the last century, Sweden, I believe, was amazingly law-abiding by modern standards. Meanwhile, in America, there was prohibition of alcohol and a frightening rise in organised crime. Sweden has a long history as an advanced country. I seem to remember from my history lessons long ago that Gustavus Adolphus once cut a swathe through Europe. At that time, the USA was wild country. Swedish furniture at the time of Napoleon was sophisticated. The country was religious and had strong friendly societies, on which their social security system was originally based.

America, in contrast, had slavery, civil war, mass migration, fights with the native Indians and so on and on. These countries were not exactly two identical substances in a scientific experiment which could be treated in different ways to measure the effect of the two different treatments.

I would suggest that the welfare states in both countries have been extremely damaging and have, in both cases, led to increases in crime far beyond what they would otherwise have been. But it is clear that the two countries started from different places. Sweden, in living memory, was outstandingly law-abiding. That has been changing. As David Popenoe remarked in "Disturbing the Nest" (1988),

"The very rapid and far-reaching social and cultural changes in Sweden from 1960 to 1985 have not been without their problematic social consequences. Once a society where following the letter of the law was one of the supreme commandments, for example, Sweden has witnessed in that period a widespread increase in lawbreaking. According to most international comparisons, the lawbreaking in Sweden may seem miniscule (and it is certainly not violent) but by previous Swedish standards it is immense."

He goes on to say that "during the 1970s, Sweden had the highest number of reported thefts per 100,000 of the population of the 75 countries included in Interpol's international crime statistics"

And again, "the most dramatic postwar development in lawbreaking, however, has been the enormous increase injuvenile delinquency" (pages 157/168 ppaperback edition).

I would suggest that the impact of a welfare state is not instant. The culture and history of a country are powerful. But the welfare states take effect with a lag. Saints are not turned into violent criminals overnight. The rise in juvenile delinquency Popenoe referred to could well be a marker for the future.

I am tempted to say a lot more on this absorbing and important subject but I must try to stop myself responding, I have so much else of a completely different nature that I have to do.

I will just add the general point that the USA has much more of a welfare state than most people realise. Forty per cent of healthcare is paid for by the government. Most children go to government schools (and I have interviewed refugees from such schools who reported the guns, knives and promiscuity there). Meanwhile Sweden, though it certainly has a major welfare state, is not as thoroughly statist as many assume. It has some independent hospitals, independent schools and, of course, some major businesses such as Ericsson.

Again I thank you for the points you have made and apologise if I do not make further responses. Over the past two years or so, I have spent a lot of time answering objections to my book and most of them have been depressingly lacking in content. It has been a positive relief to engage in a worthwhile debate.

Posted by: James Bartholomew at December 7, 2006 05:25 AM

One small point - John Hudson used incarceration rates per capita as a measure of crime. While I appreciate that such figures are more reliable, given differing crime recording methods in different jurisdictions, they do imply, for example, that the USA has a lower crime rate than the UK, whereas the opposite is true.

I would disagree with James' statement that "the state, in addition to being a bad supplier of welfare, tends to be a bad administrator of everything else it does, too. This applies to all aspects of criminal justice including investigation of crimes, prosecution, the courts and punishment."

The state managed criminal justice pretty well for the 100 years to 1955 or thereabouts. It managed education pretty well too. The factors which have led to the collapse of state efficiency in such areas are cultural ones, not structural ones. Some powers should be reserved to the state - defence and criminal justice being among the first.

If you are suggesting that such functions should be privatised ... well. Let's just quote the great economist PJ O'Rourke.

"never let the people with all the money and the people with all the guns be the same people"

Posted by: a at December 7, 2006 11:15 AM

I believe that comparisons between countries is useless in this case. While it may well be true that welfare state in Sweden has not caused a rise in crime, that is of no probative use at all when considering the British position. The two countries have completely different histories and their welfare states are not the same. The matter is so complex that any reference to Sweden, or any other jursidicition, in an argument about the effect of the welfare state in britian is totally facile.

Posted by: Peter L from Oz at December 8, 2006 03:04 AM

1. The Swedish crime rate _has_ risen with the rise in 'welfare' spending.

2. The bulk of prisoners in the US are there for drug 'crimes' -- eg, possession of marijuana etc. That is, they're _not_ there for 'serious' offenses against people or their property.

Posted by: Sudha Shenoy at December 9, 2006 04:04 PM

Using the imprisonment statistics could lead to a distortion based around the will of the State in question to imprison people. In the US they are willing to imprison for fairly minor offences, where as in the UK it seems like the judicery will do everything it can to avoid people being sent to prison (partly because they are full). Perhaps using homocide statistics would be better as they are harder for other factors to distort.

Posted by: chris at December 11, 2006 12:56 PM

I think you will find that the developed countries with the lowest levels of crime are Singapore and Japan which have lower social spending that Sweden etc. This is probably because they have stricter punishments.

Posted by: David at December 11, 2006 03:21 PM