More information on “the Hedonic Regression case”

Click here to view table.

Summary

The claims involved the valuation of premiums under the Leasehold Reform Housing and Urban Development Act 1993. The UTLC listed three applications for the specific purpose of considering the validity of a hedonic regression model (“the Parthenia model”) to determine leasehold relativity. The Tribunal indicated the approach to be adopted for assessing relativity for different lease lengths and commented on the use of published relativity graphs.

Facts

The three cases each involved applications under section 48 of the 1993 Act to determine the amount payable for a new lease. The claims related to flats in Central London. They were specially listed to decide the appropriateness of using the Parthenia model to determine the value of the existing lease on the statutory assumption that the Act conferred no right to acquire any interest in any premises containing the lessee’s flat or to acquire any new lease. The use of an earlier version of the Parthenia model had already been rejected in Kosta v Carnwath [2014] UKUT 0319 (LC), where the UTLC adopted an average of a number of published relativity graphs.

The UTLC considered the issues in some detail over a 9-day hearing and its decision ran to some 170 paragraphs together with a further 36 pages of appendices.

Decision

The main conclusion is that the UTLC rejected the use of the Parthenia model for the purpose of determining relativity. The model is incompatible with market evidence, and in the words of the UTLC, “the Parthenia model is a clock which strikes 13.” There are a number of other technical criticisms which were set out in detail in Appendix B to the UTLC’s decision.

The UTLC commented on various published relativity graphs in Appendix B to its decision.

  1. First, it considered the Savills “Enfranchiseable” graphs of leases with rights under the 1993 Act. There are two versions (with different methodologies). At the valuation dates, the Savills 2002 Enfranchisable graph was in common use. Savills’s 2015 Enfranchsisable graph was a considerable improvement on this, and (if various technical criticisms can be overcome) this graph could in time properly replace the 2002 version.
  2. Second, it considered the various graphs of relativity without rights:
    1. The Tribunal “gained no assistance from” graphs produced by the College of Estate Management and John D Wood.
    2. It did not find the graph produced by WA Ellis “to be useful”.
    3. The graphs produced by Charles Boston, Cluttons and Knight Frank were “all based on small sample sizes and each had particular problems”.
    4. At the valuation dates, a prospective purchaser would not have taken an average relativity from all the graphs (as in Kosta).
    5. The Gerald Eve graph was the “industry standard” and prospective purchasers acquiring leases would most likely have referred to the Gerald Eve graph on the valuation date.

At paras 163-169 of its decision, the UTLC set out those matters which it considered might be useful in determining relativity in future cases.

  1. The Parthenia model should not be put forward in future cases.
  2. Tribunals and experts must focus on the state of the market at the valuation date. That market may be influenced by published graphs, but it is not open to a party to suggest the market was historically badly informed or operating illogically or inappropriately.
  3. The market might perform differently in future (e.g. as a result of tribunal decisions) and any changed market circumstances would have to be taken into account.
  4. In many cases there will have been a ‘real world’ market transaction in respect of the existing lease on or around the valuation date. If so, that transaction is likely to be “a very useful starting point for determining the value of the existing lease without rights”. It would normally be possible for an expert valuer to express an opinion about the deduction for ‘Act’ rights on the statutory hypothesis (for example, the UTLC accepted a deduction of 10% for Act rights made by the landlord’s valuer in Aaron v Wellcome for a lease with 41.32 years unexpired).
  5. Where there is no market transaction, “valuers will need to consider more than one approach”. One possible method is to use the most reliable graph of relativities for leases without rights. Another is to use a graph of Act world relativities to ascertain the relative value and then make a deduction from that value to reflect the value of rights. If the two methods throw up different figures, the valuer will need to weigh the strengths and weaknesses of the two methods.
  6. For the FHVP, the Savills Enfranchiseable (2015) graph is likely to be beneficial (see above).

Comment

This is an exceptionally important decision. Although not specifically stated to be a ‘guideline’ case, valuers should follow the approach to leasehold relativity suggested by the UTLC. The primary method of valuation suggested (i.e. a real world trascation adjusted for Act rights) is in a sense an easy one, although opinions may well differ on the value of ‘Act rights’. The secondary method (i.e. a comparison of graphs) relies on a selection of the appropriate graph as well as an assessement of ‘Act rights’. Some graphs were praised (especially Savills Enfrancshisable and Gerald Eve) but the Tribunal did not specifically reject the use of others. However, the Tribunal did not favour the commonly used method of simply averaging graphs of relativity.