House of Commons - Trade and Industry

Select Committee on Trade and Industry Written Evidence

APPENDIX 20

Memorandum by Morgan Stanley

SUMMARY

We are publishing two pieces of research on the tenanted pub industry following the UK government's announcement it is going to investigate the nature of the beer "tie". Our qualitative research is a postal survey of 13,000 tenants in conjunction with trade paper the Morning Advertiser. Our quantitative research analyses the financial results of quoted pub company operators Enterprise Inns and Punch Taverns to assess exactly how much cash their tenants are paying/forfeiting to operate their pubs.

1. The survey makes fairly gloomy reading. It was posted and completed before the investigation was announced, yet 624 tenants still responded (a high 5% response rate), over 400 of them with Enterprise and Punch. Our three main conclusions are as follows (fuller summary on page 4, complete findings and methodology on page 22):

— Poor relationship with landlords. Around half our respondents rated as poor or very poor their landlords' help in developing their business, their commitment to investing in the pub, and their fairness as a business partner.

the average was

under 10%

— An unexpectedly-high £17,000 was spent on capex in our respondents' pubs last year, equivalent to around 10% of sales, and well above what the Pubcos say needs to be spent annually (£5,000-10,000). Most of this was financed by the tenant rather than the pubco.

The survey is subject to significant bias we believe. After all, when asked, which tenant is not going to want a rent cut? As an indication of this, nearly one-fifth plan to leave the pub trade, and one-quarter regret going into the trade. This cannot be wholly representative of the industry or else there will be no one left to run our pubs in a few years!

2. Quantitative assessment. From the data we conclude: The average tenant is paying 7% of asset value to operate it annually, in-line with average commercial property rates. This is the net impact of rent, foregone margin from being tied to high beer prices, and shared machine income, but after netting off the accommodation benefit. We show this figure has been constant over time:

after

— No evidence Pubcos are generating supernormal profits. Indeed, Punch and Enterprise's return on capital are not exactly sparkling at 7-8% (albeit above-WACC).

— Not much evidence of "stress". Pubcos quote a variety of key performance indicators (bad debts, etc), and on all counts the figures are both small and shrinking.

The discrepancy between the two pieces of research can really only be explained by the survey picking up a disproportionate number of weak pubs. After all some tenants genuinely are struggling, but this can often be the tenant's own fault (as Enterprise suggests). We cannot ignore the financial analysis that the "average" tenant seems to be faring OK, with some at the other end of the bell curve clearly doing very well indeed. After all, it would be surprising to say the least that Enterprise and Punch had run their business so aggressively when this is an entirely symbiotic relationship.

Jamie Rollo

Leisure Analyst

Morgan Stanley

EVIDENCE FROM OUR PUB SURVEY

We have conducted a survey of 13,000 pub lessees and tenants in conjunction with pub trade paper the Morning Advertiser. We received 624 completed surveys (a statistically significant 4.8% response rate), of which two-thirds are tenants/lessees of quoted operators Punch Taverns (including Pubmaster) and Enterprise Inns (including Unique). The survey was completed before the DTI announcement that a Select Committee would investigate the tenanted pub industry. We reproduce our full methodology and findings in Appendix 2.

Readers should remember there is an element of bias in this survey, as in most surveys. When asked if they would like to pay less rent, what tenant isn't going to answer yes? Similarly, 46% of respondents regret going into the pub trade a little or more, 23% intend to sell their lease in the next year (or one-third of the Enterprise and Punch respondents) and 21% are looking to leave the trade in the next year. This cannot be representative of the wider industry or else there would be no one left to run our pubs in a few years! We are left with the inescapable conclusion that the survey is skewed towards the weaker end of the market. Readers should bear this in mind, particularly as 94% of respondents thought that the OFT should investigate the industry.

Our five key conclusions are as follows.

1. The relationship between tenants and Pubcos is not great, to say the least. Indeed, the data make fairly depressing reading, with the problems apparently stemming right from the outset of the tenancy or lease contract being drawn up. Exhibit 1 summarises some of the responses to questions asked

— One-quarter of respondents said the initial contract they took out was explained to them in an unsatisfactory manner by landlords (could there be a case for "mis-selling" here?). One-third of respondents said that disclosure of information was poor or very poor. We would argue that the tenants are partly to blame, as only 59% took legal advice. Note these figures are little changed from levels in the 2003 survey, despite greater focus by pubcos on offering greater flexibility and improved support. This could reflect the lag as contracts only gradually come up for renewal

— Over half (53%) the landlords apparently did not make clear what investment they would make on the pub, and only half actually spent what they committed to spending (few put it in writing)

— Over half (55%) of respondents rated their pubco as poor or very poor in helping them generally develop their business. This includes very bad scores on advice on developing the wine and food offerings (the principal growth drivers in the industry)

— Nearly half (45%) rated their landlords' fairness as a business partner as poor or very poor. This is close to last year's proportion (47%).

On a more positive note, the relationship with the Business Development Manager (BDM) has improved. Around one-third of respondents see their BDM once a month, but nearly half do so only once every six months, and 15% do so annually or less often. That nearly two-thirds see their BDM semi-annually or less is not great, and pubcos might need to add extra resources here. Still, respondents generally rate their personal relationship with the BDM between satisfactory and good, well above the relationship with their corporate pubco itself, and there has been a slight improvement here year on year. Punch's score has improved significantly since the last survey; Greene King, while still above average, has slipped.

2. A full 53% of respondents think their rent is too high, a similar proportion to last year (54%). This is perhaps not surprising given the previous feedback (and turkeys do not vote for Christmas). Respondents' rent averages around 13% of turnover, which matches the recent data given by the pubcos.

However, when asked what reduction would make rent fair the average was under 10% (among those who deemed rent to be unfair in the previous question). We show the split in Exhibit 2, which shows that, of those who stated a view on the matter, nearly 80% would deem a rent reduction of 10% or less as fair. Indeed, this reduction is less than the 13% average reduction wanted a year ago, perhaps reflecting the fact that Punch and Enterprise are starting to offer more rent reductions. So, these respondents may think things are slightly unfair, but not terribly unfair.

This implies a worst-case scenario of only a 5% average rent reduction at the Pubcos (that is, half of pubs seeing a 10% rent reduction). This would imply a 5% EPS reduction, given rent constitutes half of EBIT, and EBIT covers fixed interest costs only two times.

3. We were surprised at how much capital investment is going into these pubs. On average, over £17,000 was spent on repairs and refurbishment in the past 12 months, and tenants expect nearly £21,000 to be spent this year. This is equivalent to around 10% of sales, or 5% of asset value, on our estimates. This is well above what the pubcos say needs to be spent annually (£5,000-10,000). That most of the money is financed by the tenant rather than the pubco suggests that tenants are earning enough to pay for the infrastructure of the pubs. Within this number, spend at Punch pubs is above average (£19,300), with that at Greene King pubs well below (£7,000).

Exhibit 4

2004 TENANTED PUB SURVEY— ARE YOU MAKING MORE MONEY THAN LAST YEAR (%)?


Total	Enterprise	Punch	Unique	Pubmaster	Greene King

Yes	33	34	29	30	26	19
No	60	65	62	61	62	76
Not stated	7	1	9	9	12	5

Source: Morning Advertiser/Morgan Stanley 2004 Tenanted Pub Survey.

4. A mixed outlook from the tenants. Over one-quarter (26%) of respondents feel more confident about their pub's business prospects compared with last year, but 36% feel less so. Enterprise's and Unique's tenants feel less confident than average (44% feel less confident versus last year). One-third of respondents say they are making more money than last year, but 60% say they are not. 58% of respondents say that complying with government red tape has reduced their profits in the past 12 months, while 38% say it has not. On average, respondents say they have lost £3,917 in profits due to red tape.

5. There are some significant directional changes within the landlords

— Punch Taverns has made most improvements year on year, but remains one of the lowest scorers. Punch once again is deemed to be the least fair licensee (23% of respondents voted for it, a rise from 18% last year). However, this seems somewhat at odds with the feedback from its own tenants, who signal Punch is making the best efforts to be a better business partner.

— Enterprise also experienced some deterioration in perception, with 16% of total respondents deeming it to the least fair landlord, up from 9% last year. Still, this is lower than its 21% share of respondents in the survey, implying that it is relatively well-perceived.

— Greene King remains one of the highest-rated landlords, but did experience significant slippage in certain areas. In terms of outlook, its figures are significantly lower than average, with only 19% of respondents saying they are making more money and 76% saying they are not.

EVIDENCE FROM THE FINANCIAL DATA

It is hard to conclude from the hard data below that lessees are paying too much for their pubs, that they are struggling, that there are cheaper alternatives to enter the pub industry, or that pub companies are making supernormal returns.

How tenanted pub companies work. Pub companies own most of their pubs on a freehold basis, and rent them to tenants and lessees through a variety of contracts. These agreements generally lead to pubcos generating income from three different sources

— Rent, which is fixed at the outset of each tenancy or lease, and based on the open market value of the pub as agreed between both parties. Usually, rent is paid monthly in advance, is indexed to inflation annually and is subject to review after a certain period (increasingly, pubcos are removing the `upwards only' nature of the agreement)

— Sales of beer and other wet products bought exclusively from the pubco. This "tie" results in lessees paying an above-market price for their beer, and the pubco's purchasing power means it takes a significant discount from the brewer. In return, lessees pay a below-market rent to the Pubco

— Income from leisure machines, derived from a profit-sharing arrangement with some lessees, as well as from commission from machine operators for siting their product. In addition, there is a small element of income from items such as cellar service and buildings insurance, which some contracts stipulate are conducted with the pubco.

We believe the best way to consider this is to ignore the PR/noise and focus on the hard numbers. Exactly how much are tenants paying in relation to the value of the pub, how much income are they generating for themselves, are there any signs of "stress" (bad debts, etc), and are pubcos showing signs of generating supernormal returns?

1. How much are tenants paying in relation to the value of the pub? A tenancy or lease is a low capital cost way for individuals to enter the pub market. Rather than pay £400,000-500,000 for a freehold, they can take out an agreement with an annual rent of £25,000-30,000. Of course, tenants invariably end up paying more than this via the tie for beer and other products, and we have used Enterprise Inns' and Punch Taverns' financial statements to assess exactly what tenants are paying in total.

In 2005 we estimate the average yield to the pub companies will be around 7% (we use 2005 as 2004 is distorted by acquisitions). Exhibit 6 explains how we reach this figure. Note that this includes £8,000 (around £150 per week) to reflect the saving to the tenants for not having to pay for accommodation. We believe our figures are on the conservative side (for example, we assume all AWP income pubcos generate is effectively foregone by lessees).

Exhibit 6

TENANTS EFFECTIVELY PAY A 7% OR SO YIELD TO RENT THEIR PUBS

Note: For full supporting data (on a total basis, as well as over time), please refer to the back of the report.

Source: Morgan Stanley Research estimates.

This 7% effective yield the lessees pay is in line with commercial market yields. Hillier-Parker's retail property index yield is 7.8% currently. Note the following, though

— Rents are actually set in relation to turnover, which runs at around £180-250,000 for an average tenanted pub (with Punch and Enterprise towards the upper end of this range). After deducting their accommodation benefit, this implies tenants are paying around 14-16% of sales. This is in line with the EU's guidelines for a free-of-tie rental assessment being set at 15% of sales

— This all assumes asset values are correct. If pub balance sheet values are overstated, tenants are clearly paying more. However, there is no evidence of major write-offs when pubs are sold. Punch normally males a small capital gain, and Enterprise makes gains and losses when tend to net out.

— Enterprise seems to be charging a little more than Punch. However, Punch has a less conservative asset valuation policy (it includes a 20% or so lotting premium, without which we estimate its yields would be closer to 8-8.5%—on the high side, but not unreasonable). This explains why Punch appears to generate similar yields to Enterprise, but much higher profit per unit of beer volume sold (see our analysis in Appendix 1).

Furthermore, there is no evidence from these companies that they have been squeezing up yields over time. Exhibit 7 shows the same calculation taken back to 2002 for Punch and Enterprise. For the past three years, yields have been remarkably static, at around 7%.

2. How much income are tenants and lessees generating for themselves? Enterprise quotes its lessees as generating £37,000 equivalent income per annum and Unique's as making £40,000 per annum. Punch and Wolverhampton & Dudley both quote £35,000. These figures include the non-cash accommodation benefit referred to above. We show our own estimate of a shadow P&L for the average lessee (work we initially published in our report Tenanted Pubs: Not Yet Drinking Up Time, July 30, 2003).

Exhibit 8

P&L OF AN AVERAGE TENANTED PUB


(£'000)

Beer	117.9
W/S/M	39.3
Food	30.0
Machine income	20.8
Total sales	208.0
Sales per week (£)	4,000
Beer	(69.0)
W/S/M	(14.5)
Food	(12.0)
Machine income	(16.6)
Total COGS	(112.2)
Gross profit	95.8
Margin (%)	46.1
Direct pub costs (inc repairs allowance)	(31.2)
Incremental staff costs	(7.8)
Profit per finance costs	56.8
Rent to Pubco	(28.1)
Interest on F&F, etc	(1.1)
Total operating costs	(29.1)
Pub income (before tenant salary)	27.7
Rental cover	2.0
Add back accommodation benefit	8.0
Equivalent income	£35.7

Note: We have attempted to build a basic model to show the financial position of the average tenant (not that there is one). This is based on an average pub in Enterprise's and Punch's estates, selling 234 barrels of beer a year, paying rent of £28,000 (£120 per barrel) and buying at a price that gives the pub owner a net discount of £150. We also use industry-average gross margins (around 50% for a tied tenant), assume direct costs are around 15% of sales (electricity, rates, hygiene and other sundries) and include wages for one temporary helper.

Source: Morgan Stanley Research estimates.

In other words, take-home pay seems to be around £26-32,000 per annum. This matches the "half divisible profit" rule of thumb that was used to set rent in the past (and Punch and Enterprise generate around £25,000-30,000 rent per pub). This is not much on a per hour basis, given it usually pays for a husband and wife team, often working 70 hours each (that is, £3.50-4.00 per hour, well below the £4.50 minimum wage). However, remember there is no mortgage or rent to pay for accommodation, and that tax rates are effectively lower for the self-employed.

On this basis, many lessees seem to be making a comfortable living.

3. Could tenants operate in the market more cheaply? Pubcos argue it is much easier, cheaper and safer renting a pub from them than buying one (see Exhibit 9). The upfront cash requirement is minimal, so the return on capital can be higher. Availability of pubs is also high, with 600 for rent by Enterprise and Punch alone. And, if things go badly wrong, the tenant loses his deposit, whereas the freeholder could lose much of his bank loan.

4. Are there any signs of "stress" (bad debts, etc)? Both Enterprise and Punch have been very forthright with these `key performance indicators in the last six months (these figures were not even quoted a year ago). Exhibit 10 shows the important measures for Punch, Enterprise and Unique

— A substantive agreement (that is, not a tenancy at will or probationary tenancy) can be an indication of a higher-quality tenant or lessee. Around 7% of Enterprise's combined estate and 10% of Punch's are non-substantive. This includes pubs both companies currently have for sale.

— Bad debts are 0.2% of sales for Enterprise and 0.4-0.5% for Punch, both very small. Note that this figure can be hard to measure, as many debts are owed for years and only taken as "bad" when a provision is made to write them off.

— Rent concessions offer tenants struggling to pay rent a few months' grace. These equate to less than 0.5% of Enterprise's rent roll and 1.6% of Punch's.

— Rent reviews sent to arbitration last year numbered zero at Enterprise, while the five at Unique were all settled in its favour and Punch had under five.

— Dilapidation claims refer to properties that have not seen enough investment. At Enterprise, £800,000 of such claims equate to 0.2% of the company's enlarged asset base.

— Finally, Punch has five times as many applicants on its database as it has pubs to fill, and Enterprise has two times as many. Punch actually has 8,000 applicants on its database, but reports only those who have completed a competency questionnaire and credit check.

Furthermore, these figures seem to have improved over the last couple of years (for example, Enterprise's bad debts were 0.3% of sales last year).

Exhibit 9

GETTING INTO THE PUB INDUSTRY—COST OF RENTING VERSUS BUYING

Assume a Pub Worth £450,000 with a rent of £26,000 and barrelage of 210 barrels p.s.

DO I BUY OR LEASE?

Source: Punch Taverns.

Exhibit 10

MAJOR PUBCOS' KEY PERFORMANCE INDICATORS


	Enterprise Inns	Unique	Combined	Punch Taverns

Pubs on short-term agreements (%)	91	95	93	90
Bad debts as % turnover	0.2	0.2	0.2	0.4-0.5
Value of rent concessions (£mn)	(0.5)	(0.6)	(1.1)	(2.0)
Rent concessions as % of rent roll (%)	0.4	0.5	0.4	1.6
Rent reviews	410	263	673	90
Rent reviews sent to arbitration	0	5*	5
Dilapidation claims (£mn)	(0.7)	(0.1)	(0.8)	N/AV
Applicants on database	691	164	855	1,600
Pubs to let	195	99	294	300**

N/AV = Not available. * All settled in Unique's favour. ** 450 including assignment. Note: figures for Punch exclude Pubmaster.

Source: Company data, Morgan Stanley Research.

5. Are pubcos showing signs of generating supernormal returns? Finally, we have looked closely at the return on capital being generated by the pub companies over time. Exhibit 11 shows the trend over the last few years, together with our forecasts to 2007 (this is an adjusted post-tax, post-cash maintenance capex return). We draw the following conclusions:

— Returns have been falling, if anything. As we have discussed before, Enterprise has seen returns decline over the last five years, though mainly due to it paying higher multiples for ever-larger (but higher-quality) pub estates

— The absolute level of return on capital is not great, at 7-8%. This is above the low WACCs of 6.5-7.0%, on our estimates, but mainly because WACCs have dropped faster than returns, as companies have been gearing up with their acquisitions (a finite process).

Conclusion. It is difficult to conclude from the hard data that lessees are paying too much for their pubs, that they are struggling or that pub companies are making supernormal returns.

OUR CONCLUSION

We were surprised by the announcement of the review, even though an investigation was one of our "Top 5 Surprises for 2004" (see our report Outlook for 2004; and Five Possible Surprises, 9 December 2003).

We would be even more surprised if the Committee found a major problem, given that the beer tie has survived numerous official scrutinies over the past decade, and is a well established part of UK pub and brewing history. Indeed, the OFT rejected a request from the Federation of Small Businesses (FSB) to investigate pubcos in 2002 (the Committee's Chairman refutes allegations that he did not know about this before launching his inquiry).

Without wishing to second-guess the Committee, our conclusions are as follows:

— It is difficult to conclude from the hard data that there is a big problem. We find no evidence that pubcos are making supernormal returns, that the average tenant is paying more than an open-market property yield, that the average tenant is struggling, or that there are many cheaper alternatives to run a pub.

— The survey results make gloomy reading, and the pubcos clearly have a lot of work to do to improve relationships with lessees. However, while over half respondents believe their rent is unfair, most of these tenants would deem a 10% reduction as fair. Even ignoring the natural bias behind this sort of survey, this does not suggest to us that tenants are suffering terribly.

— It looks as if a minority of tenants are being very vocal. We concur with the pubcos that some tenants are simply bad retailers, and that, if the turnover of the pub has fallen since the rent was set, this is not necessarily the fault of the pubco. Punch's decile analysis shows that the bottom three deciles of its estate (where the tail of poor performers lies) account for 10-12% of total EBITDA.

— Tenants should not ignore the non-cash benefit they obtain through free accommodation and other perks.

— This is a symbiotic relationship. It would be surprising if the pubcos were ripping out short-term profits to the long-term detriment of their business. It is not in a pub company's interest to over-rent a pub, as if this leads to bankruptcy or non-payment of rent, whereby both sides lose.

The Committee may still recommend some remedies, if for nothing else than to be seen to act and to protect vulnerable tenants from themselves. Possible recommendations could include:

— Require that tenants and lessees taking out new contracts seek legal advice. Perhaps pubcos should be responsible for ensuring this, or even paying for this

— Launch a Code of Conduct for all parties to abide by, rather as in the supermarket sector, to deal with complaints

— Remove "upward only" rent reviews. Contracts have improved—"upward only" rent reviews were removed in 1998 for Enterprise and Punch's new agreements have done away with them (although the old Bass and Vanguard leases contain them). The government hoped its Code of Conduct for Commercial Leases would lead landlords to remove these terms, but many smaller companies still have them. The Code also deals with insurance terms, and encourages openness of negotiations between landlord and tenant. The Office of the Deputy Prime Minister said that, if the voluntary Code was not adopted by 2004, it would sponsor new laws to make "upward only" rent reviews illegal, and force landlords to offer break clauses and short leases to those who want them. Recent surveys suggest that over 90% of leases taken out still have "upward only" reviews, and few Pubcos have changed their leases. The larger pubcos (Enterprise and Punch) appear to be in a minority. We can see a situation where contract terms are made entirely flexible as regards length of lease, rent/beer price mix, machine supply, etc.

— Remove the obligation to acquire ancillary services. Many tenants complain that their agreement requires them to insure the building through the pubco, for a cost that can be many multiples higher than were they to insure independently. Likewise, pubcos charge for cellar services, when this can involve little more than checking that the tenant is not buying out of the tie. The Committee may decide to require pubcos to clarify the hidden cost of these ancillary services.

— More transparent beer pricing. Pubcos buy their beer for around half the price their tenants do, and the difference ("discount" or "commission") generates around half their income stream. Tenants are required to buy at the brewer's list price of, say, £300 per barrel, whereas freehouse operators could get a discount of £70 per barrel, and the pubco a discount of up to £140 per barrel. The discounts have reached the point where tenants are paying more for beer than Wetherspoon customers are paying for a pint over the bar. Of course, this beer income helps offset the below-market rent in the tenancy. However, we also believe that the pubcos have been generating higher discounts over the years without passing much back to the tenant (see Exhibit 14). Many contracts rebate to the pubco well over half of price increases via higher commission, which the tenant effectively ends up funding. If the Committee demands pubcos make this pricing structure transparent, we think there could be a surge in "buying out".

Finally, the Committee may, of course, find enough evidence to refer the issue to the OFT. If this is the case, we would still not expect many onerous undertakings from the ultimate conclusion, but investors would need to prepare for a rocky ride.



© Parliamentary copyright 2005	Prepared 14 March 2005