Monthly Archives: March 2015

5 in 6 Men? A Response to Ally Fogg

Prison Population by Gender

Figure 1: Prison population in England and Wales

[Added May 2016: An update on these issues can be found here]

Ally Fogg has recently published a criticism of my blog piece UK Prisoners – The Genders Compared. I respond to his comments here. In my article I claimed that, if men were treated like women in the UK criminal justice system, then 5 out of every 6 men in prison would not be there.

Let me make one thing clear: I have no intention of dying in a ditch to defend the 5-in-6 estimate. I would not, as Mr Fogg asserts, be embarrassed to be proved wrong. I am not a feminist or a politician that I cannot admit error, if such be the case. I would, in fact, be happy to be proved wrong. The salient word here is “proved”. I would be happy to be proved wrong because it would mean that someone had examined this matter with more rigour than I have been able to do – in which case our knowledge and understanding would be advanced. This is the scientific mentality. That said, I will address each of Ally Fogg’s points.

I shall make use of two Ministry of Justice documents in particular: Statistics on Women and the Criminal Justice System 2011 and Statistics on Women and the Criminal Justice System 2013, henceforth referred to as Refs.1 and 2 respectively.

Firstly let me summarise my case. I observe that, for people convicted, men are given a custodial sentence 3.4 times more frequently than women, and, moreover, their sentence is, on average, a factor 1.64 times that of women. A statement along these lines will not be contentious, I hope. In fact, these numbers are probably worse. For example, Ref.2 Figure 5.01 gives the custody rate for men and women to be 10.3% and 2.5% respectively, a disparity factor of 4.1, larger than my estimate of 3.4. It also gives the average custodial sentences as 16.2 and 8.9 months respectively, making men’s sentences a factor 1.82 times that of women, again larger than my estimated 1.64. These disparity factors explain why there are more than 20 times as many men in prison as women, despite the fact that only slightly more than 3 times as many men are convicted of crimes per year. As I say, these observations should not be contentious.

The issue, though, is whether this disparity between the genders is just, i.e., whether there are good reasons for men receiving harsher sentences, or whether it is simply gender bias. It is perfectly possible, in principle, that the disparity could be just. For example, if a far larger proportion of men committed heinous crimes, such as severe violence, whilst most women’s crimes were minor offenses like petty shoplifting, then it could be that men were simply getting their just deserts. I attempted to address this question by breaking down both the above factors into the different categories of crime. The outcome was, I claimed, that broadly similar factors are found in virtually all crime categories. On this basis I concluded that the disparities were gender bias since, if a man and a woman commit a crime in the same category, the man can expect to be treated more harshly to a degree reflected, on average, by these disparity factors.

I now address Mr Fogg’s points, which will be quoted in italics.

(A) huge problem with Collin’s analysis is that he doesn’t allow for the fact that men and women tend to be prosecuted for different types of crime.

As explained above, I have specifically addressed the issue of different crime categories. Mr Fogg asserts that men tend to commit more heinous crimes (as a proportion) than women. I have two responses to this: the first is that it is not true (I address the sexual assault issue below), and the second is that it would not matter anyway since broadly similar disparity factors apply across all crime categories. In my blog post I took the simpler course of appealing to the latter argument, thus avoiding a discussion of the relative frequency between the genders of the different categories of crime.

However, the blog post was actually a précis of a more detailed analysis here. In that original I also attempted an analysis of the proportions of the prison population convicted for the various crime categories, comparing the genders  (see Fig.4.3 on page 42). It was only crudely done, but my tentative conclusion was that the similarity of the offense ‘spectrum’ between the genders was more striking than the differences. This further encouraged me not to concentrate on this aspect.

I can now add to the evidence underwriting this view. Ref.1 Figure 4.02, reproduced as Figure 2 below, shows the proportions of males and females sentenced (i.e., convicted) in various crime categories. I suggest the similarities are indeed more striking than the differences (note the similarity in the ‘violence against the person’ category).

Contrary to Mr Fogg’s assertion, a very similar pattern of offending between the sexes is again displayed by Ref.2 Fig. 4.02 which shows the proportions of males and females arrested in the various crime categories. (Note that such people were not necessarily tried in court, and, if so, were not necessarily found guilty). Again the spectrum of offending is strikingly similar between the sexes, including in the category ‘violence against the person’.

Figure 2: From Ref.1 (2011)


Figure 3: From Ref.2 (2013)


I wonder how many people know that the most common reason for a woman to be arrested is for violence?

Or that, combining violence and sexual offences, the proportion of men and women arrested is essentially the same (34.4% and 34.9%, ref.2 Table 3.03).

Of course, these Figures are percentages of the totals for each sex, not absolute numbers. The ratio of females to males “proceeded against” is the same as the ratio of convictions, namely 25% to 75% (Ref.2 Figure 5.01). Consequently, the rough equality in the proportion of men and women arrested or convicted for violence+sexual offences implies that there is one women convicted of these offences for every three men.

The first point (William Collins) misses is that the single biggest influence on whether a defendant is sent to prison (and for how long) is not their gender, or even the category of offence, but their offending history. Someone with 15 or more previous convictions is nearly five times as likely to receive a custodial sentence as someone with no previous convictions. Of all people processed in the justice system, women constitute 27% of first offenders, but only 14% of repeat offenders. In other words, while men represent about 73% of offenders in the criminal justice system, 86% of people processed for repeat offences are male…..I would suggest that it is at least likely that amongst the most chronic recidivists, an even higher proportion is male.

In principle this is a valid objection. My simple broad-brush analysis takes no account of offending history, but judges will certainly take this into account in passing sentence. It has long been a contention in some quarters that women are treated more harshly in the criminal justice system because they are more likely to be sent to prison on a first conviction. In his speech to the House of Commons on 16 October 2012, Philip Davies, MP for Shipley, was at pains to point out that this is the opposite of the truth. On the basis of Ref.1 he noted that, “Of sentenced first-time offenders, a greater percentage of males were sentenced to immediate custody than females (29% compared with 17%), which has been the case in each year since 2005“. This picture is confirmed for 2013 by Ref.2 Figure 6.02, which shows that, for first time offenders, men are nearly twice as likely to be sent to prison as women (28% versus 15%) whilst women are more often given suspended sentences or a conditional discharge, see Figure 4.

Figure 4: From Ref.2


However, Mr Fogg’s point is not quite that. He asserts that men are far more likely to be serious recidivists and thus to fairly attract more severe punishment. But what are the facts?

I believe that Mr Fogg’s claim that only 14% of repeat offenders are women relates to data on being cautioned or convicted combined, since I can find that percentage in Ref.2 Table A.01 and also in the “1 in 7” figure on page 51, both of which relate to this combined category. However, what we are interested in is, for people who are sentenced (i.e., convicted), what are the relevant proportions of repeat offending and how does this influence the likelihood of imprisonment?

Ref.2 page 53 indicates that the large majority of both sexes who were sentenced for an indictable offence in 2013 were repeat offenders, namely 86% of women and 91% of men. Moreover, whilst male offenders sentenced for an indictable offence were more likely to have 15 or more previous sanctions, the difference between men and women in this respect is slight: 37% of men and 30% of women. So Mr Fogg’s claim that chronic recidivists are overwhelmingly men is not true.

In respect of the sentences received by such recidivists, Ref.2 page 53 tells us that, “In 2013, the most common disposal for offenders convicted of an indictable offence with 15 or more previous sanctions was immediate custody for both males and females, reflecting that repeat offenders are more likely to get an immediate custodial sentence. A higher proportion of males (40%) with 15 or more previous sanctions received an immediate custodial sentence compared with females (31%)“. So, as anticipated, repeat offenders of either sex are more likely to be imprisoned, but there is again a disparity against males even for a comparable degree of recidivism.

Perhaps most critically for the analysis, Collins assumes that men and women committing the same category of offence are committing the same type of offence. This is an enormous, and entirely unsupported assumption. Violence against the person includes everything from minor assaults to mass murder. Theft could mean shoplifting a jar of babyfood or an enormous gold bullion heist. Fraud could be a failure to declare to the Job Centre a few quid earned for babysitting or it could be a billion pound financial scam.

[Aside: Mr Fogg naughtily leads the reader by using baby-care issues as illustrations of minor offences. Please replace “shoplifting a jar of babyfood” with “a starving homeless man stealing a loaf of bread”. Incidentally, shoplifting is recorded explicitly as such – and men are “far more likely” than women to receive an immediate custodial sentence for shoplifting – Ref.2 page 15.]

It is true that I have implicitly assumed that, on average, offences committed by the two sexes within a given category are equally serious. It is also true that I have offered no support for this. The assumption is made simply on the basis that, in the absence of contrary evidence, this is a reasonable working assumption. Mr Fogg is right that many crime categories cover an enormous range of seriousness. But, as I say in the original post, why should men’s offences for, say, “theft from a vehicle”, or “false accounting” or “fraud” or “causing death by reckless driving”, etc., be any more heinous than women’s on average? In the absence of specific data, the assumption of equity on this matter is reasonable. Mr Fogg disagrees, he says,

Knowing what we do about different patterns of offending by gender, it is utterly nonsensical to assume that the types of offences typically being committed by men and women within the categories are comparable.

I don’t know what Mr Fogg thinks he knows about the differing patterns of offending, but to my mind Figures 2 and 3, above, indicate very similar patterns of offending. Against this background I suggest that my assumption is reasonable, if unsupported, rather than “utterly nonsensical”. The alternative would be to arbitrarily, and without justification, attribute the entire “factor of six” apparent gender disparity to differences of crime severity within each individual crime category, an issue on which…..

We simply do not have the data.

…..and hence cannot be critically examined. This does not necessarily mean that it is untrue. There are limits to what an amateur blogger with access only to publicly available information can do in a few days of effort. To do justice to the topic would require a PhD level of investigation, and access to far more detailed data. The true indictment of our society is that no one seems to have done it.

Of course, the very purpose of Refs.1 and 2 is supposed to be to address potential gender discrimination. Specifically, there is a legal obligation under Section 95 of the Criminal Justice Act 1991 as follows, “The Secretary of State shall in each year publish such information as he considers expedient for the purpose… of facilitating the performance of those engaged in the administration of justice to avoid discriminating against any persons on the ground of race or sex or any other improper ground…”. But Refs.1 and 2 do not make any conclusion on the question of discrimination. Who is charged with the duty of reassuring us on this matter?

There are aspects of these documents which I find worrisome. Why, for example, does the title refer only to “women and the criminal justice system”? And on page 18 of Ref.2 we are told that, “additions are based on suggestions from members of the Women’s Independent Advisory Group and reflect the needs of users of the report“. The additions referred to are new sections on mitigating and aggravating factors which are deemed to have an impact on sentencing disparities. That it was felt necessary to add these sections (between the 2011 and 2013 editions, and hence after Philip Davies’s speech in Parliament) suggests an unacknowledged recognition that a potential case of bias needed to be addressed. That the additions are “based on suggestions” from a women’s group is hardly reassuring when the matter in hand is potential gender bias.

The new sections on mitigating and aggravating factors are based on the Crown Court Sentencing Survey. This is a survey given to all judges sitting in the Crown Court, for them to complete every time a sentence is passed. The forms record details of the factors taken into account by the sentencer in deciding the sentence given to an offender. The survey is voluntary and has a response rate of around 60%. The survey is only completed for those offenders sentenced at the Crown Court, where the types of offences committed are more serious than those that go through the Magistrates’ Court. As the large majority of offenders are sentenced at the Magistrates’ Court (around 97% of females, compared to around 91% of males) the survey covers only a small fraction of sentencing. Superficially the survey provides extra information on how sentences were derived.

The very clear trend is for more mitigation to be applied to women and for more aggravating factors to be applied to men. This may be taken as justifying sentencing disparity. But since most of the categories are value judgments, and the survey is (literally) a tick-box exercise undertaken after the judge has already made his decision, it is inevitable that the survey should align with the trend of sentencing. To take just one example, females received mitigation compared to males based on “physical or mental illness”. Given that 90% of prisoners have a mental disorder it is hard to rationalise this. Similarly, women receive mitigation compared to men based on “remorse” and “good character”. Interestingly, and in contrast, where an issue is factual rather than a value judgement, there can be little difference. For example, for assault and public order offences, men and women are subject to an equal aggravation penalty based on “the actual or threatened use of a weapon”.

Ally Fogg’s last point relates to sexual offences,

Around one in six of the male prison population (around 11,000 excluding remand prisoners) is serving a sentence for a sexual offence. Among women, the figure is less than one in 50. Why? Because there are 56 men prosecuted for sexual offences for every one woman. Meanwhile sexual offences are the category where convicted (people) are more likely to be imprisoned than any other, with around 60% of convicted offenders (whether male or female, incidentally, there is no major difference here) receiving immediate custodial sentences. A good illustration of how far awry Collins’ calculations have taken him is that without any prima facie evidence of biased sentencing in sexual offences, those 11,000 sexual offenders would still be there if male sexual offenders were sentenced identically to female sexual offenders. And yet he assumes that only only 13,000 men should be in prison in total.

Mr Fogg has a point. I confess that I was rather beguiled by the fact that sexual offences account for only about 3% of men’s convictions each year (see Figure 2) and paid it little mind. But my broad brush analysis does not really apply to this exceptional issue. Mr Fogg is correct to point out that somewhere around 10,000 to 11,000 of the male prison population (currently 81,700) are inside for sexual offences, i.e., about 13% – far greater than the conviction rate due to the greater likelihood of imprisonment and probably a longer sentence.

But just a moment. My thesis is that men are treated far more harshly in the criminal justice system than women. Is Ally Fogg really suggesting that sexual offences, of all things, provide a counter-example to this claim? For heavens sake, is this not the area which is the most egregious example of different treatment of the genders? The discrimination in this case lies in the very fact that “56 men are prosecuted for sexual offences for every one woman“. But if men and women were truly subject to the same justice, would there really only be one female offender for every 56 men? The discrimination in this case goes much deeper than the criminal justice system. It is endemic in our society and prevents female offence in this category being recognised – at least, that is my contention.

What evidence is there for the sexual offending of females being far higher than the 56-to-1 prosecution ratio would suggest? I’ll not attempt to summarise it all here. I suspect Mr Fogg knows all about it anyway. But as one example, Michelle Elliott estimates that perhaps around 20% of paedophiles are women (see also the review here, on page 100). Or take the issue of female school teachers sexually exploiting their pupils. In a 2009 dataset I have to hand, under “sexual activity with child under 13” and  “sexual activity with child under 16” no women were sentenced to imprisonment. Under “abuse of trust- sexual offences ” and “gross indecency with children” and “familial sexual offences (incest)” there were a total of just 4 women imprisoned. And yet one can list hundreds of cases of female teachers in the USA being convicted for sexually exploiting their pupils. Yes, the US population is greater, but only about five times greater. It seems likely, though I have no proof, that women in the UK having sex with under-aged boys is under-recognised and under-prosecuted. And then there is the quagmire of adult female on adult male sexual abuse. Again, shrouded in mystery and controversy, but almost certainly far more widespread than is recognised either in society as a whole or in the criminal justice system in particular.

So, do I claim that my 5-in-6 figure is correct to the 9 decimal place accuracy of quantum field theory? No, of course not. But it remains my best estimate until such a time as someone does a demonstrably better job. I hope they do. The real issue is that there is discrimination and it is of substantial magnitude. Although giving it a precise numerical value is not strictly the point, there’s nothing quite like a specific number for making people sit up and take notice. My analysis is certainly crude but I hope I have demonstrated that it does not deserve to be written off as “arrant nonsense”.

Finally, having responded to all Mr Fogg’s points, I’d like to close by returning to Figure 1. It shows a roughly 8-fold increase in male prisoners over the last 75 years.


Crime in general, and violent crime in particular, has been falling for at least the last 10 years, and perhaps the last 20 years (see here, page 30). So why has the male prison population continued its steep increase?

Discrimination is not the only possible answer. But the 8-fold increase does align crudely with my identified factor of 6, and it has occurred over the timescale in which denigration of men has become widespread and acceptable. If it is not discrimination within the criminal justice system then it surely must be a consequence of the societal changes which have gone hand in hand with the aforesaid denigration: family breakdown, the marginalisation of fathers and the loss of male investment in society. So, to end on a point of agreement with Mr Fogg,

Prison, for both men and women, is an expensive, ineffective and inhumane anachronism in a civilised society. Pretty much every inmate detained is a testament to multiple failures in social policy, social care, education, welfare, mental health and addiction services. There are innumerable strong arguments for reform of sentencing and penal policies.

Indeed. But the anger which Mr Fogg so dislikes in the MRM arises because the likes of the Corston Report promulgate the view that only women are deserving of such compassion – when it is so very obvious that it is specifically men who need it more.

If Mr Fogg wishes to leave a comment here I will afford him the privilege of having the last word. I will not indulge in comment Ping-Pong.