The efficient packing of children

Housing benefit is a UK benefit which helps low income tenants pay their rent. Entitlement is, reasonably enough, set out in the housing benefit regulations. About a decade ago I was looking at the part of these regulations which governs the size of property it’s considered reasonable for the claimant to occupy.

13D
(3) The claimant shall be entitled to one bedroom for each of the following categories of occupier (and each occupier shall come within the first category only which applies to him)—
(a) a couple (within the meaning of Part 7 of the Act);
(b) a person who is not a child;
(c) two children of the same sex;
(d) two children who are less than 10 years old;
(e) a child.

It looked pretty straightforward, but after writing several test cases I realized there was a problem where the claimant:

  • has three children of the same sex; and
  • one child of the opposite sex; and
  • one of the children of the same sex is aged at least 10 and the others are not

The law prescribes an order in which to allocate rooms. But where we have odd numbers of the same sex, it does not tell us how we select the children to pair off under 13D(3)(c). If we pair off the older same-sex children first, we will be left with two children aged under ten, who can share a room under 13D(3)(d). If we choose to pair up the younger same-sex children first, we are left with two children who cannot share a room – they qualify for a room each under 13D(3)(e).

From a programmer’s viewpoint this ambiguity feels highly unsatisfactory. And we might well take as the moral of this slight tale that we should all be trying to execute our regulations before they come into force, rather than leaving ambiguities to be deliberated upon later by tribunals.

But it’s possible to read the regulation another way. Since the housing benefit regulations do not prescribe a way to pair off the children, the decision maker can take the broader family circumstances into account. Whichever pairing is chosen will comply with the law. This would give leeway, say, to pair up young twins and give separate rooms to teenage siblings, rather than forcing teenagers to share with toddlers. In another case where children were closer in age, the decision maker might award one less room.

Unfortunately for this optimistic reading, the corresponding provision for universal credit (an integrated benefit which will mostly replace housing benefit for working-age claimants) closes off any discretion on the part of the decision maker:

(2) A member of the extended benefit unit to whom two or more of the descriptions in sub-paragraph (1) apply is to be allotted to whichever description results in the renter being entitled to the fewest bedrooms.

This removes the ambiguity, at the expense of discretion.

I don’t know if the drafters of the original regulation gave any thought to the leeway they were possibly allowing. It may be unlikely, given that it involves a relatively uncommon circumstance. But I do wonder if thinking of law as code might tend to favour rigidity over discretion. This is not necessarily a bad thing of course – discretion can be abused, and in any case can be explicitly provided for. Perhaps the most useful thing that coding can offer law is a set of practices (testing, encapsulation, patterns) rather than a metaphor.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>