Last night Enfield Council’s overview & scrutiny committee met to consider whether the cabinet’s decision to implement private rental sector housing licensing should be confirmed, referred back to the cabinet or referred to the full council. Valid questions from some members of the committee, and from the public, were ignored by the chair of the committee, by officials, and by the cabinet member for housing. New evidence, published here in the public interest, shows why the council blocked the uncomfortable questions.
I have previously written about the flawed statistical basis and Enfield Council’s failure to consider alternatives to its plans to license landlords and properties in the borough’s private rented sector. The scrutiny committee should have considered just one question last night, before deciding on a course of action:
Is there sufficient evidence that there is a link between anti-social behaviour and the private rented sector to make a designation for additional and selective licensing under the Housing Act 2004?
The discussion went into other topics, such as the consultation process and whether this will be just a revenue-making scheme. The cabinet member for housing, Cllr Ahmet Oykener, asserted that the scheme would not make a profit for the council. Cllrs Alan Sitkin and George Savva said they were fighting for those people who live in terrible conditions because of rogue landlords.
These are all important matters but entirely peripheral to the question before the committee. The key question – whether the legal test for designation of additional and selective licensing had been met – was put to both Cllr Oykener and Ray James (director of health, housing and adult social care) but was not answered. The committee’s chair, Cllr Toby Simon, who also sits as a magistrate, even shouted down a local resident who wanted to make sure that the question was answered. What was that question?
What advice did Enfield Council receive from legal counsel, on what was an adequate correlation between anti-social behaviour and the private rented sector, before proceeding with designation of additional and selective licensing under the Housing Act 2004?
Cllr Edward Smith had asked for a pause on any decision, pending the findings of a current review of government policy in this area, and evaluations of other schemes. Cllr Terry Neville, who is not a member of the committee but had led the call-in of the decision, asked that the matter be referred to the full council, which would have met after the forthcoming local elections.
After many heated exchanges the matter was put to a committee vote and was passed on a majority. The cabinet’s decision to go ahead with additional and selective licensing was confirmed.
The question of whether a scrutiny committee can adequately scrutinise the executive cabinet when it has a majority of the same political colour is one we can ponder some other time. Last night important questions went unanswered. Serious charges of information being withheld by council officers were ignored. Councillors Oykener and Simon shouted down the local electors, whom they purport to represent, and threatened them with removal from the meeting.
The dodgy statistical report
The report commissioned by Enfield Council from Cass Business School: Professor Les Mayhew consultancy, Mayhew Harper Associates / Neighbourhood Knowledge Management, undertook analysis in two steps:
- to try and predict which homes were in the private rented sector;
- to analyse any associations between the private rented sector and anti-social behaviour.
We have already critiqued that work, based on what was published by Enfield Council.
My own experience is that analysis is only as good as the questions asked, the assumptions made, the conclusions reached and the presentation made. If any one of those is flawed, then there is a problem. The report was heavily caveated but maybe not strongly enough. I certainly do not agree that there is sufficient correlation to make the case for this policy.
Cllr Sitkin went to great lengths to state Professor Mayhew’s expertise. But even big names make (occasional) mistakes, whether by action or omission. He would do well to acquaint himself with the case of Harvard professors Reinhart and Rogoff, whose simple spreadsheet error was used to support austerity measures in economies with high debt.
What Enfield Council didn’t want the public to know
This website has obtained a document which shows that the report produced by Professor Les Mayhew gave rise to misleading conclusions, despite heavy caveats. The letter from Sally McTernan, the council’s assistant director for community housing services, is addressed to a local resident who raised questions about the statistical report. This document is disclosed here, in the public interest [pdf, redactions, highlighting and notes in red are mine; the document question numbers are cross-referenced below].
We can leave it to the consciences of Mayhew and council staff as to why this analysis was omitted from the report. Here is what we know from the document but that the cabinet member for housing and senior council staff, including chief executive Rob Leak, were unwilling to answer:
- Quoted directly: “NKM analysis does not conclusively prove that residents of private rented properties are the main cause of anti-social behaviour in Enfield.” [Q1]
- The models that Mayhew used to try and establish a link between anti-social behaviour and the private rented sector had a goodness-of-fit of 55.0% for single family dwellings and just 0.35% for houses in multiple occupation (HMOs). In the case of single family dwellings, this means that even if the model is valid (and this could be tested if data were made available), 45% of the variation is caused by something else. This was after they had removed outliers at Trent Park and Chase Farm Hospital. Their result would have been worse had they left them in.There is absolutely no correlation between ASB and HMOs. [Q2]
- Mayhew confirms that “HMOs show no relationship with ASB”. [Q26] Remember those maps that the council used to justify their claim? A significant chart – the one showing no relationship whatsoever – was omitted from the report. Would you pass a local law which didn’t even have a 1% link with the problem being addressed?
- We already knew that the ASB could only be linked to addresses for about 5% of cases. Prof Mayhew states that ASB rates were higher for single family PRS dwellings but this is based on a tiny fraction of the cases. [Q3] He assumed “association by proximity”. [Q21, Q27] Would he take the blame for someone else’s crime just because they were in the same place at the same time?
- “ASB tends to concentrate in the east and south” of the borough. These are the parts to the east of the A10 Great Cambridge Road and to the south of the A406 North Circular Road. And these are the parts flagged up in my previous critique as being areas of deprivation, which Mayhew argues is not a good indicator because it is four years old. But hang on, he is using data which is three years old… [Q4, Q26] 24
- Assumptions were discussed with council staff. The huge assumption made was that Newham is similar to Enfield. [Q5] What other assumptions were made after consulting Enfield staff? What effect did this have on the actual or perceived independence of the work?
- There was no validation work in Enfield. Neither council staff nor MHA/NKM undertook visits to confirm the accuracy of the predictive model. An accuracy of 90% is claimed. [Q6] This 90% is based on visits that were a combination of the model and “other prior information” (translation: tip-offs). The validation work in Newham was incomplete, in any case. If we design a test for apples, we would check how good it is at identifying the apples, as well as how good it is at rejecting the pears. In this case, apart from the tip-offs issue contaminating their result, they only looked at how good they were at identifying the PRS properties. They didn’t confirm that houses predicted as outside the rental sector were indeed owner-occupied or social housing. Furthermore, they make a heroic assumption that the “risk factors” for Newham – the characteristics used to predict whether a property is in the private rented sector – apply to Enfield.
- As with so many types of analyses, it’s often more important what has been left out than what has been put in. The last sentence is quoted directly from Mayhew’s follow-up answers [Q28]. It hints at a pre-determined outcome from what should have been an exploratory data analysis to look at differences between different tenure types in the borough, if any.
“No work was done on the social housing sector because that was not the aim of the project, although it would be possible to do so for comparative purposes, in addition the identity of social landlords is already known and social housing is already subject to regulation.”
Let the facts speak for themselves. Was democracy demonstrated or defiled?
You can also look at other posts in my Enfield Council series, including a review of the statistics (needs assessment) and the legal position, with alternatives they could have followed (policy options).
If you found this useful, please click on “like”, share it, leave a comment below, sign up for email updates (no registration required) or get in touch about what I should look at next.
Image attribution: Titian’s “Tarquin and Lucretia”, Fitzwilliam Museum