Index Linked Rent Reviews (Updated) 0 Comments

Index Linked Rent Reviews (Updated)

Some years ago, I published a blog post on my website on the subject of index linked rent reviews in leases of commercial property.

Since my earlier post, much has changed in the world of indexation and, in addition, the subject appears to have become newly topical in the context of long leases of residential property reserving ground rents.  The time is ripe for an update.

This post is not primarily concerned with the issue of residential ground rents referred to above, but it’s hard to ignore it and I will comment relatively briefly.

Residential ground leases

At one time (until at least the middle of the 20th Century) ground rents under leases of houses and flats were fixed throughout the term at a peppercorn or a nominal number of pounds. Over probably the last 20 or 30 years, however, the practice of landlords (usually developers) granting leases of new build houses and flats has been to provide for some degree of escalation of the initially reserved ground rent.  Whether this is driven by a wish for income protection or income maximisation (or greed?) is an open question.

One increasingly popular model is to increase the ground rent by a fixed amount at periodic intervals; an extreme example is the doubling of the ground rent every 10 years. This has not unnaturally attracted a great deal of censorious comment as the unfortunate tenant can, over a period, find himself liable to pay a very substantial sum and this inevitably will affect (in extreme cases destroy) the value and/or marketability of his property. Mortgage lenders have started to introduce rules as to the maximum degree of escalation which is permissible if they are to lend on the security of the leasehold property and the government promised action; the exact form of any legislation is unclear but reference has been made to a statutory requirement that ground rents under residential leases should at no time exceed the traditional peppercorn (or a nominal sum of money).

My primary concern in this post is to address the question of whether under any circumstances indexation by reference to any supposed cost of living-based index is an appropriate way of reviewing rental values in property of any kind. Critics of the proposed legislation in relation to residential property have, however, made the point that the great majority of lessees can (in the case of a house) enfranchise by acquiring the freehold under the Leasehold Reform Act 1967 or (in the case of a flat) claim an extended lease under the Leasehold Reform, Housing and Urban Development Act 1993.

I would be interested in the views of valuers on how a potentially very high ground rent would affect these rights. In the case of enfranchisement, my understanding is that the starting point in determining the price payable (before the division of marriage value) is the combination of the capitalised value of the landlord’s ground rent and the value of his reversion (the reversion being treated as the present freehold vacant possession value deferred for the unexpired lease term). The capitalised value of the ground rent will presumably rise with the ground rent – so there is not much comfort here for the tenant. The landlord’s reversionary value will, as I understand it, not be reduced (as would the leasehold market value) by a very high ground rent as what is to be valued is the freehold vacant possession value deferred at the appropriate rate.

Image - Index Linked Rent ReviewsThe situation is probably not much better for the tenant of a flat in the case of a lease extension. Although the usual expression is “lease extension”, what the tenant is entitled to claim is a new lease for the unexpired term of the existing lease plus 90 years. The rent payable under the extended lease (with immediate effect) will be a peppercorn. Unfortunately for the tenant, however, he must also pay a premium equalling the amount by which the open market value of the landlord’s interest in the flat is reduced by reason of the grant of the lease extension. Clearly much is going to depend here on the stage at which the tenant exercises his right. The extent of the decrease in the landlord’s value is clearly going to be related to the amount of the ground rent he is losing, so presumably the impact of an increase in ground rent will be reduced if there are many years to go before the increase(s) take effect (again, I am ignoring marriage value).  Finally on this topic, if the proposed legislation is to be retrospective (and reduce to zero ground rents payable under thousands of existing leases), how are landlords to be compensated?  There is nothing illegal about these practices, so confiscation without compensation would be wrong.  An idea of the scale of the problem can be gleaned from The Guardian post here. If just one of a group of (according to The Guardian) 85 companies has freeholds valued at £274+ millions, the scale (and potential cost) becomes apparent – and that is the tip of the iceberg as many thousands more freeholds must be held by investment and development companies, individuals and institutions.  Again, real estate advisors may have researched the market and be able to come up with an overall estimate.

I can now return to my main proposition, which is primarily concerned with the review of commercial rack rents.

The purpose of rent reviews

Moving back to the main subject, the starting point must be to look at the reasons why landlords would consider indexation by reference to cost of living indices as an alternative to the traditional “open market rental value” review basis. It is worth noting at this point that some landlords try and have it both ways by seeking to review rents to the greater of the indexed figure or the open market rental value, but the problems canvassed below will apply anyway, because the indexed figure still has to be calculated. Wherever the indexed figure has to be calculated, it is not usually true indexation, as the majority of landlords will insist on rent review being on an “upwards only” basis whatever the link. But to my mind, the key point is that any reliance on a cost of living-based index defeats the fundamental purpose of rent reviews. It is often said that rent reviews are included in leases to protect the landlord from the effects of inflation, but I do not think that is wholly accurate. I think the best judicial statement of the purpose of a rent review comes from Sir Nicholas Browne-Wilkinson VC in the case of British Gas Corporation v Universities Superannuation Scheme Limited [1986] 1 ALL ER 978. The Vice Chancellor said:

“…there is really no dispute that the general purpose of a provision for rent review is to enable the landlord to obtain from time to time the market rental which the premises would command if let on the same terms on the open market at the review dates. The purpose is to reflect the changes in the value of money and real increases in the value of property during a long term.”

An indexation-based review may (though see below) protect the purchasing power of the landlord’s rent, but it does not do what a rent review clearly should do – i.e. bring the rent into line (almost invariably on an “upwards only” basis) with rental values prevailing at the time of review. After all, the original rent would almost invariably have been based on open market values (usually by reference to comparable lettings in the market); there is no way in which an original rent could be based on an index. To that extent, indexation alone really defeats the purpose of rent review (again, some landlords manage to get away with a “higher of” basis but as I’ve said, that still remains problematic in that it requires the indexation calculation to be made).

I said above that an indexation-based review may protect the purchasing power of the landlord’s rent but in fact there is some doubt about that. Institute of Fiscal Studies LogoWriting in The Times on Tuesday, March 21st 2017, Paul Johnson, Director of the Institute for Fiscal Studies, commented (in the context of the decision of the Office for National Statistics to publish three measures of inflation – see below) as follows:

“What inflation as we measure it does not mean is change in the cost of living – the change in the amount of money I need to be as well off this year as I was last year. It does not, for example, take account of the act that if the price of wine skyrockets, I may buy less wine and more beer. In this sense, all our measures probably overstate the increase in the cost of living. Inflation simply means the change in prices faced by consumers. 

“The word “simply” though is misleading. How does one understand “change in price” let alone measure it, when completely new goods – smartphones for instance – become available? What should one do when the quality of the same goods – cars, televisions, computers – changes almost beyond recognition over time? How can a single figure account for the different spending patterns of millions of households?”

That aside, indices based on an across-the-board sample of retail prices have always been political. This means that they are liable to be manipulated by governments for one reason or another (mostly to do with the yield on gilts). This can easily be demonstrated by reference to the history of the relevant indices, of which there are currently three: RPI, CPI and CPIH (though the RPI is no longer a national statistic – see below).

The history

Historically, an interim Retail Prices Index was introduced in 1947 and made official in 1956, but the first official Index of consumer prices was actually introduced in 1914. Even before that, there were measurements of consumer price inflation; see  the Office for National Statistics (ONS) publication “History of and Differences between the Consumer Prices Index and Retail Prices Index” (which can be downloaded here).

A new RPI Advisory Committee was convened in the early 1980’s. This produced a wide-ranging report which led to many changes in the definition, scope and coverage, the treatment of subsidies and discounts, and the treatment of owner/occupiers’ housing costs. There were further Advisory Committees during the 1980’s and 1990’s. Between 1995 and 2008, a number of changes were made in the calculation methodology and new arrangements for the governance of what was then the RPI were established by the Statistics and Registration Services Act 2007.


The CPI has a much shorter history than the RPI. It was first introduced in 1996 as the Harmonised Index of Consumer Prices (HICP). HICPs were developed across the European Union for the purposes of assessing whether prospective members of the European Monetary Union would pass the inflation convergence criteria, and of acting as a measure of inflation used by the European Central Bank to assess price stability in the Euro area. What the EU was looking for was a consistent measure of inflation across Europe in order to make reliable comparisons between EU Member States – not generally possible using the RPI and the Historic National Prices Indices in other Member States due to differences in Index coverage and construction.

In December 2003, the national statistician decided that the name of the UK version of the HICP would be changed to the Consumer Prices Index in all national statistics publications. The ONS document lists 9 changes which had, by the time of its publication, been made to the CPI since its introduction.

The point of including this historical material is to demonstrate that both the RPI and the CPI already have a history of periodic change, and to provide a background for a brief look at the difference between the RPI and the CPI, and how they impact on the use of either Index for the purposes of rent review.

Again, to save space, I refer readers to the ONS document but the principle and most well-known differences between the two Indices is that the CPI excludes the following housing-related items which are included in the RPI:

*  mortgage interest payments;
*  house depreciation;
*  buildings insurance and ground rent;
*  house transaction costs such as estate agents’ fees, home buyers’ survey costs and conveyancing charges;
*  council tax.

Shopping BasketThe CPI also excludes television and road fund licenses on the same basis as it excludes council tax (i.e. they are considered to be taxes and do not constitute household final monetary consumption expenditure under national accounting rules).

To confuse matters further, there is now a “Consumer Prices Index including owner/occupiers’ housing costs” (CPIH). The CPIH is now a national statistic, but I am not proposing to examine it in detail as it is of recent origin and I have yet to see it used for rent review.  In broad terms, the CPIH should be closer to the RPI than the CPI – but see below as to the different methods of calculation.

The choice for landlords wishing to index rent has traditionally been between the RPI and the CPI (with the CPIH also now available) and I think I have said enough to make it clear why landlords tend to favour the RPI – the RPI simply tends to rise at a higher rate than the CPI.

This trend is magnified by the way in which the RPI is calculated. The formula used in calculating RPI is known as the Carli formula. This is known by economists as “the formula effect”. The RPI and CPI use different formulae to aggregate individual price quotes into Elementary Aggregation indices (EAs) for individual items. The technicalities are beyond the scope of this post (and me!) but as I understand it the CPI predominantly uses the geometric mean (also known as the Jevons Index), supplemented in most cases by a particular form of arithmetic mean known as the Ratio of Average Prices (the Dutot index). The RPI predominantly uses a combination of the Dutot and the arithmetic average of price relatives (the Carli index). The difference which this creates is well illustrated by a piece in The Times newspaper for 1st August 2017 by Tom Knowles. I quote:

Tom Knowles (The Times) Article on Retail Price Index - 1st August 2017“If you want to know how big a difference using either the retail prices index and the consumer prices index makes to calculating inflation, look inside your wardrobe… .  Using the same set of data, the Office for National Statistics found that the price of clothes under CPI had increased by a respectable 15% between 2010 and 2016. Under RPI, the same data showed prices rising by 80%.

“This came down to the Carli formula used to measure RPI. When tweaking the way it collected clothing prices, the ONS discovered in 2010 that this formula, first used in the UK in 1947 and the measure of headline rate of inflation until 2003, was seriously flawed.

“The national response would have been to switch to the more common Jevons formula, but there was an issue: there was £407 billion of index-linked Government bonds whose returns are calculated using RPI. Any change would lead to the Bank of England having to decide what to do about gilt holders who could see their returns drop, and that then would have been an issue for the Chancellor. 

“Instead, the ONS decided to discredit the RPI but change nothing. The Government did the same in 2013, after the Debt Management Office had decided against switching new index-linked gilts to rise with CPI. Never mind the ordinary tax payers paying £2 billion more of interest every year than would be the case if the RPI had been corrected”.

Mr Knowles’ analysis refers to the impact of all this on employers who have final-salary schemes moving in line with RPI and to the fact that RPI is still used to calculate the interest on student loans and rail fare rises. He comments: “The excuses for doing so seem slim” and concludes “The measure may be “flawed” but it is clearly profitable”.

The same justification might be used to criticise landlords using RPI – although it is worth noting that in my experience, in cases where rent review clauses in commercial leases have required the calculation of the “best of” open market value or RPI-linked figure, the open market value has tended to win except in very poor property markets (the existence of which is no doubt the reason for the inclusion of the indexation formula).

The problem of change in the method of calculation

The comments above illustrate what I said earlier: the indices are affected by political considerations and the basis on which either index is calculated can be changed (for example, he ONS did flirt with the idea of changing the basis of calculation of the RPI to Jevons, and produced the RPIJ – which for the reasons explained by Tom Knowles was quickly discontinued). The very important question arises therefore of how to draft a rent review clause so as to accommodate future changes in the method of calculation of the selected index?

UK Price Index MatrixThe normal adopted drafting approach is to provide that if the basis of calculation (itself an imprecise expression likely to cause disputes) of the selected index has changed, then the rent calculation is made on the basis of the index as it was before the change.  If the change is a simple exclusion of one item then this might work for a single rent review under a short lease, but in the case of a longer lease with multiple rent reviews (and I have seen long ground rent leases which use indexation as a basis for review) it all becomes highly problematic to the point where it may be virtually impossible for the parties (or an arbitrator or independent expert) to make the calculation.  At the very least it is likely to lead to disputes on each review date.

Examples of this approach can be found (for subscribers) in the Practical Law Company document “Index rent review clause based on RPI, clause 1.7/1.8; the Encyclopaedia of Forms and Precedents (Butterworths) Volume 22(3)A and Ross on Business Leases (Butterworths) (Precedent 28).  The PLC and Ross precedents both take this method a stage further by providing that if it becomes impossible to calculate the rent by reference to the Index (which is the situation I have just envisaged), the question is referred to an arbitrator who is to determine “a reasonable rent for the Premises”.  I do not really see how this works without some more guidance to the arbitrator on the basis on which he should look at the question; if the arbitrator is a surveyor then he will almost certainly look at market value (I am in no way criticising the authors of these precedents as they are dealing with a rent review based on indexation only without alternatives and have really gone for the only option).

I see no reason to change the conclusions at the end of my earlier post:

(a)       Linking the rent under a lease to the RPI (and I note that all of the precedents referred to above use the RPI and not the CPI) is not consistent with the fundamental purpose of rent review; and

(b)       It is unsafe for leases for any significant period of years to rely only upon the RPI or CPI as the basis for review; even providing for the higher of the indexed figure or open market value is problematic as the indexed figure will still need to be calculated and this is likely to prove difficult or impossible where the calculation basis has been changed.

If you liked this article, please share it.

This Post Has 0 Comments

Leave A Reply

Please fill out the required fields *