Description

Please see charts here:  https://www.dropbox.com/s/1gppbg7ax94lt29/Countrywide%20Writeup%2010.20.14.pdf?dl=0

[chart 1] 

Introduction (Theme)

The prevailing pitch, though revolving around Countrywide PLC (CWD.L), is perhaps a more thematic play on the UK real estate agents/lettings market. Under that theme, LSL Property Services (which was recently written up by WeighingMachine on VIC) is also a solid play with many of the same investment qualities as Countrywide. One could establish a mixed position in both names to diversify some of the idiosyncratic risks (such as LSL losing a surveying contract).

Investment Thesis

Countrywide PLC is the largest real estate agent in the UK, with ~1,350 branches spanning 46 local brands. The Company operates under 5 core functions, all directly tied to the health of the UK housing market (and more directly, to the number of housing transactions. This number was just shy of 1 million in England and Wales in 2013, versus long term averages of ~1.4 million). Countrywide’s core business is real estate brokerage– where the Company’s agents sell homes and collects commissions of ~2% (varies by market). Supplementing this function are the “sell-throughs” – mortgage origination (financing), conveyance, and surveying. In aggregate, these value-add businesses make up 25% of revenues and 31% of pre-corporate EBITDA. The fastest growing segment in recent history been Countrywide’s lettings business (rental) – the division has performed well as it is countercyclical to housing construction/transactions, and has been augmented by a series of acquisitions

We believe Countrywide is an interesting investment due to 1) absolute valuation at ~12x 2014E[1] P/E with housing transactions still ~15% below 30 year averages, 2) capex light business which translates to high FCF, and a business strategy to return ~75% of FCF to shareholders 3) reinvestment opportunities in “rolling up” lettings agencies that have historically been done at ~5-6x EBITDA (20-25% targeted return after 2 years), 4) incremental EBITDA margins across all sectors at 40+%, and 5) structural differences between the UK and US market that make market disruptions less practical.

The risks to the business reside predominantly in the health of the UK markets, unexpected increase in professional indemnity claims, price competition from resurgence of smaller estate agents, and the rise of internet portals like Zoopla and Rightmove. In addition, there are external political risks, evident in the April mortgage review (which puts limits on mortgage loans), and further modifications that may come in other forms.

The Business

Countrywide is the largest real estate agency in the UK, with an estimated market share of ~6-7% and ~50% of off-channel mortgage distributions. The Company makes money by helping sellers find a buyer, and on average charges a 2% commission – along with the transaction, Countrywide has ~55% sell through of value-added services such as mortgage financing or conveyance.

While real estate agents have localized specialty (Countrywide and LSL both maintain numerous local brands), Countrywide’s scale allows it to promote all listed homes across its network, and to that end has allowed it to create its own online channel, Propertywide.co.uk. generated ~1/3 of all Countrywide leads last year (the remaining split roughly between Zoopla and Rightmove, the two largest real estate advertising channels in the UK w/ near complete market share). Real estate and London and Premier together accounted for ~50% of revenues and ~40% of pre-corporate EBITDA. 

[Chart 2, 3]

In addition to the core business of real estate and lettings brokerage, Countrywide also provides logical value-add (read: “sell through”) services such as surveying, financing and conveyancing. In aggregate, these 3 components of the business (which CWD believes its real estate division provides ~50-60% of the volume) generated ~£34 million of EBITDA, or roughly 28% of the Company’s aggregate pre-corporate EBITDA.

In short summary:

• Surveying is the undertaking of diligence to appraise a home (for mortgage approvals, etc.). CWD does 1 in 4 UK mortgage valuations. The business was capacity constrained industry-wide (LSL and Countrywide all invested in new employees towards the end of 2013, which dampened margins, but should lead to new business in 2014 and beyond. CWD is targeting 25% capacity increase by end of 2014). While the business has high ROIC, it is also responsible for the personal indemnification expenses that CWD incurred throughout the boom. One hopes they are more diligent now…
• Conveyance is the transfer of legal titles from one person to another. This industry is super fragmented with an estimated 8,400 firms in the UK, a 1/3 which do <10 a year. Why wouldn’t you convey where you buy?
• Financing – Countrywide facilitated 8.3 billion of mortgages in 2013 and increased employees 8% over 2013
• LSH is a new division that was acquired late last year and deals primarily in the commercial real estate sector. There is potential upside here, but the business is not very significant to the core of the thesis

I can’t ascertain the strengths of CWD over competitors beyond 1) its vast array of accolades, 2) its large scale and ability to generate cross-sells, 3) the Company’s market share stability. While competition is bound to increase if the real estate market continues to recover, I think it is likely that the pie expands enough for all industry players to win (at least in the short/medium term). It is important to stress that the large real estate players have all substantially reduced their fixed cost structure following the recession (down 4.5% in 2013, and likely more as the entire IT base is outsourced). Furthermore, CWD is better positioned to weather a new downturn given the rapid growth of its lettings business, which is counter-cyclical and is roughly ~30% of pre-corporate EBITDA.

1) Housing Transaction Volume Recovery

It is readily apparent why the business is highly dependent on real estate transactions – in conjunction with commissions tied to increased business, higher prices of homes also leads to higher absolute commissions (so there is somewhat of a natural hedge between volume and price).  Transactions also carry conversion rates of ~58% for financing and ~50% for conveyance. Furthermore, of the 58% that subscribe for financing, 90% also buy life/general insurance. A recovery in volume results in leverage on many aspects of the business (likewise, a weakening of the market will simultaneously hurt all over!).

We will spend a disproportionate amount of time surveying the transaction landscape. At ~13x this years estimated PE with an estimated~1.2 million of transaction volume, we think the shares offer “okay” value if transaction volumes don’t recover, but can provide significant upside if (hopefully when!) they do. We will first provide some “forecasts” to speculate on the potential recovery value, but more importantly delve into historical numbers to attempt to handicap transaction volumes under a “new normal.” On the downside, Countrywide is expensive if transaction volumes permanently fall below 1 million, but capital loss under this scenario seems to be avoidable.

There are a number of factors that suggest the future could be brighter:

• According to CML, UK transaction volumes were estimated to grow 15% to ~1.23 million for 2014 (with a ~6.5% decline to 1.15 million for 2015)[2]. Possessions will be somewhat higher in 2015 (good for CWD). This estimate was made in July 2014.
• As of 2013, residential investments as share of GDP was just below 4%, compared to pre-bubble levels of 4.5-5.0% (during 2007, this number shot up above 7%)[3]
• Longer term, per the OBR: “By 2018, we expect a similar level of transactions as in December, in line with the long-run rate of turnover of the housing stock.” (~1.5 million transactions)[4]
• Based on projections from the 2008 census, the actual number of households is 375,000 below long term forecasts. Assuming this takes 5 years to recover, this would add ~75-80k transactions each year.[5]
• [Chart 4] 

• In addition, OBR further estimates that housing prices will increase by 30% between 2018 and 2019[6] (weighted towards 2014/2015), bringing housing prices across the UK on par with pre-crises levels in real terms.  YoY pricing forecasts seem to fall and stabilize around 4% in outer years. It is important to remember that drivers of the business include both transaction volumes as well as price (and sell through).

The entire sector has sold off recently despite these forecasts on account of commentary on the heated markets (primarily London), as well as curtailments of mortgage lending limits (April MMR). While Foxton tempered the market with reference to a weaker second half, the data through July 2014 doesn’t seem to suggest an overwhelming affect from the mortgage caps[7]:

[Chart 6] 

In addition, we think the focus on short term transaction volumes take away from the longer term picture (focusing on a 2018+ one). Below we depict the 40 year history of transactions since 1971[8], along with a more selective chart to show recent improvements into 2014.

[Chart 7, 8] 

It is true that ownership of homes have skyrocketed from 1951 to 2011, but per the ONS[9], home ownership in England+Wales in 1981 was 58%. This means that between 1981 and the peak ownership in 2001 at 69%, the ~10% increase in home ownership should not be accounted for as normal transactions (not part of turnover). In those 20 years, there were 31,364 transactions. Assuming 10% of 2011 households were newly initiated, some 2,340 transactions were new transactions. This still leads to ~29 million “turnover” transactions in those 20 years, or on average ~1,450 per year in England and Wales.

Forecasts are all fine and dandy, but let’s double check to see what those numbers are implying (and whether they’re reasonable). Per the ONS, there were 23.4 million households in England+Wales in 2011, of which 64% (or 15 million) were owner occupied, and the remainder were lettings. At the current rate of transaction volumes of ~969,000 a year (2013), and with the conservative assumption that the 8.3 million “lettings” are static, the average turnover of a home is ~15.5 years. Assuming that “lettings” are turned over every 25 years, this implies that of the 969k transactions, 332k are lettings, and the average turnover is ~23.5 years for owner occupied properties.

It’s quite difficult to make assumptions on the stay of owners in their homes, but let’s see what we can find:

• Per a US NAHB article by Dr. Emrath, around 50% of home-owners have moved out of their homes at the 15 year mark. Another dated article (1992) by Carliner suggests ~5% of owner-occupied stock turns over in a given year
• According to Zoopla/Zurich, “homeowners expect to own and live in five properties over the course of their lifetime.[10]” This might imply something between 10-20 years. Jeffries calculates that between 1978 and 2007, the average household moved once every 14 years (versus 21 years post-recession)             
• Using the 10 year census population data from 1971 to 2011, and linear approximations between decade-long estimates, the average transaction volume as % of households over the past 40 years has been 6.9%. The median has been 7.0%. This compares to current turnover of approximately ~4.1%[11].
• [Chart 9] 

• First-time home owners expect to wait 14 years before they can afford to move up the property ladder.[12] This gap compares to ~9 years between 2010-2013, and the vs. largest gap of 10 years between 1995-1999 over the last 40 years. Historically, around 700,000 transactions a year came from people “movers” – this number was estimated to be around 325,000 in 2013[13]. If we believe that global economies will recover, it is hard not to imagine this ought to revert to a lower mean

Of course, the counterarguments are:

• Average houses are now 9x average gross wage, vs 6x average gross wage 10 years ago
• Down payments are now much higher due to restrictions, as well as mortgage caps
• “second steppers” have virtually no equity to transition to larger homes post 2007, though home prices have recovered substantially across the country[14]
• Flow of foreign capital can continue to push prices up and diminish volumes/accessibility[15]

While no one can be heavily certain volumes will tread back to more normalized rates (for the thesis to work, we don’t need an aggressive recovery), I think the current pressure on lower transactions are being affected by some short-termism (fresh recollection of the crises) and manifold concentration on the quarter to quarter volumes on any slight change in mortgage requirements. While prices do seem high (specifically in London and the East, the rest of the UK is still down despite the UK average up 6.5% from peak[16]), I think this is an area that is well in the sights of government officials, whose recent actions (implementations to make non-residents pay UK property profits, multiple mortgage requirement adjustments)[17],[18] , though detrimental for price in the short run, should lead to a healthier medium and long term real estate market.

2) Capex Light Business / Shareholder Focused

In 2013, Countrywide generated pre-NWC operating cash flow of 63 million. A large majority of this was used to finance working capital, particularly provisions of 21 million.

Countrywide’s “ongoing” capex in 2013 was 15.2 million (split, roughly 60/40 on refurbishment of branches and ongoing IT costs. Another 12 million was associated with the 7 year IT transformation, which should dissipate with time). In general, the Company spends has spent ~10 million a year in capex over the last 4 years, though going forward that metric will be higher (up to 20 million through 2015).

Assuming no major changes in NWC or further provisions (steady state, with no transaction volumes growth), Countrywide did 63-15 = ~48 million of FCF in 2013, translating to around ~5% FCF yield on 2013 metrics.

[Chart 10] 

It is hard to estimate what 2014 cash flow might be given the changes in policies and a potential slowdown of the markets, but cash flow pre NWC and provisions for the first half of the year is already 22 million higher (albeit SBC continues to be far higher than anticipated). Assuming that 2H revenues are (like history suggests) ~20% higher, 2H revenues would be ~70-75 million higher than 1H. Assuming employee costs go up commensurately (around 56% of sales), but that rent, advertising, motoring stay relatively similar, then incremental margins might be ~45% over H1, which suggests 2H EBITDA could be ~30-35 million higher. Most of this should then fall through to FCF pre NWC / provisions.

Per CWD’s interim report, in absence of a major acquisition, the Company expects to return 60-70% of reported Group profits in the form of ordinary dividends and supplemental return of cash. CWD also still controls ~4% of Zoopla (worth approximately £40m), which it intends to monetize in the future (as it has with the current special diviend).

3) Capital Deployment Opportunities

The market for real estate brokers is very fragmented. The top 3 players (Countrywide, LSL, and Connells) control ~12.5% of existing branches, the top 10 16.8%, and the numbers drop off rapidly form there. ~80% of branches are owned by small/local players with <10 branches. Lettings are even more fragmented: there were an estimated 2.3 million dwellings for private rental in 2012[19] in England. Countrywide managed just 52k units in 2013, giving it a market share of just over 2%.

Given the fragmentation, Countrywide has been active in acquisitions. The Company acquired 28 lettings businesses in 2013, and continued the trend with another 16 in the first half of 2014. In conjunction, the Company continues to push its “New Starts” program (opening up lettings branches organically within its existing infrastructure. New Starts EBITDA improved to 1.4 million from -2.7 million over 2012 – margins of ~8% still a ways away from ~28% average margin of lettings segment) and has set aside 20 million to seed a property investment vehicle.

As such, this is a business with 1) very little need for capex, and a commitment to return ~75% of free cash flow to shareholders, 2) significant operating leverage based on existing infrastructure (higher tx volumes require few investments), and 3) opportunity to deploy un-returned FCF at high incremental rates. Below are the acquisitions Countrywide has done over the past couple years, per disclosures in their S-1 and interim reports:

[Chart 11]

Adjusted for commentary provided throughout the filings and on calls, acquisitions have been done at roughly 5-6x profit (I assume this is an EBIT multiple, but even EBITDA would be very accretive). The CEO further re-iterated, “we expect to make a 20% to 25% return on acquisitions in the second full year of ownership.” While this surely is unsustainable longer term[20], the opportunities still seem to be quite prevalent.

4) Strong Incremental EBITDA

Countrywide itself publishes what it believes to be adjusted EBITDA contributions (often less investments into the business and/or new business revenues) of ~45-50%. While it is somewhat aggressive to take this at heart, we can look at the incremental EBITDA margins over the last few years (these numbers don’t adjust out investments in personnel that are not yet generating revenues. As such, the ~35% incremental margins is low compared to Company derived estimates in their presentations[21]):

[Chart 12] 

CWD’s own take on incremental EBITDA margins (YTD 2014 vs YTD 2013):

[Chart 13] 

The business shows significant leverage to recovering transaction volumes, as well as increased lettings opportunities. The leverage of the business is understandable – the infrastructure / fixed costs do not change significantly, and management guided to ~50% of employee costs as variable.If volumes recover, fees will likely compress, but the model will work with significantly less incremental EBITDA margin than the Company guides (we model high teens IRR with 35% incremental margins).

Another note of interest is in the numeration report on page 50 of the annual: for long term incentive plans, 2/3 of annual awards are subject to absolute EPS targets. “25% of this part of an award will vest for EPS of 57.6 pence increasing pro rata to 100% vesting for EPS of 70.4 pence for the year ending Decmeber 31, 2016.” Now, there can surely be conflicts by driving long term incentives on EPS, but the Company appears to be expecting some strong bottom line growth (though there’s no top line guidance). Assuming a 12x-14x multiple on 57.6p to 70.4p of earnings, we are looking at a stock worth 691p to 986p based on internal targets.

5) Structural Differences

This is inherently a pretty big and somewhat obfuscated risk. In the UK, two real estate portals – Zoopla and Rightmove[22] – dominate advertising (collectively >80%, and together accounting for ~61% of online property traffic[23]).  The businesses essentially charge brokers fees to list their estates/lettings on their respective portals. Both Rightmove and Zoopla have been increasing rates year over year given their large consumer reach (brokers basically “have” to list on them). However, there are probably some limits to the extent of the two Company’s dominance:

• Real estate brokers see Rightmove and Zoopla as complete “rent” takers and as such are all responding to prevent the dominance of the companies.  In August, a group of six real estate agents, including Savills (one of UK’s largest)[24], started a non-profit portal OnTheMarket.com (Agents Mutual). The real estate brokers will be able to “pass on savings” of high listing fees[25].
• Brokers who participate can only use 1 other listing agent (hence, only Rightmove or Zoopla, but not both). This is an attempt to wrestle power away from Zoopla and Rightmove such that the real estate brokers should be able to control at least one of the large listing sites. AM currently has ~3,000 branches (vs ~16,000 for rightmove and zoopla, out of ~18-19k total in the UK) and is targeting ~5,000 by January launch. Some analysts think AM will be a non-event for the duo.
• The reality is that the hurdle for AM is much lower than Zoopla – the venture simply needs to put enough pressure on the duopoly such that the two can’t simply price up double digits year after year after year.

It is impossible to deny that Zoopla and Rightmove have dominant mind share, the most sophisticated data, and the largest community (their valuations would suggest as much). There is always risk that the status quo continues and the two continue to pass ARPA increases over its brokers. In the medium term, this risk seems manageable:

• Zoopla made 30 million from real estate broker fees in 2013. Jeffries estimates that roughly 5.3 million of that is made from major accounts like Countrywide (this seems fair, as in the UK, major players with over 50 branches represent ~19% of the market, and they also pay lower fees/account due to volume and scale). Among the major accounts as defined by Zoopla, Countrywide makes up 35-40%. This means that Countrywide probably pays roughly ~2 million to Zoopla for advertising. If Zoopla doubles its ARPA to match rightmove over time, that would only be another ~2 million of cost to Countrywide (not too big a deal).
• This is not to imply that Rightmove and Zoopla can’t increase their ARPA over time – undoubtedly they will. But it appears that the increase in costs will be manageable in the medium term (i.e. it will likely take 3-5 years for Zoopla to double and match rightmove’s ARPA). At some point in the far future the industry may get squeezed

With regards to the potential disruption of a UK “Redfin” (which would displace the actual broker itself), the risk of such a startup seems very low. The UK already has the lowest RE commission rates among developed countries (1-2%, compared to 5-6% in the United States). With such a tight spread, it becomes increasingly difficult to justify entering the market (since the potential profit pool is much smaller, and the opportunity for savings for the customer is far more limited).

As an example, Redfin started out in the U.S. in early 2000’s to disrupt the brokerage model. The Company initially charged $3,000 flat fee for selling a home and roughly ~1% to buy a home[26]. Over time, the Company realized that its customers wanted advice, and thus they had to transition to a hybrid services company and hire more real estate agents. As a result, they raised fees to 1.5% to sell a home, and reduced the “kickback” to home buyers. Even so, the old model continues to persist – as of a couple years ago, 42% of people find their homes online but 89% still use a broker, paying average fees of 5.4%[27].

Furthermore, the interaction of online-only agents, high street agents, and online advertisers is constantly evolving. Online-only agents pitch themselves as doing “everything” a high street broker does, but they also rely exclusively on Rightmove / Zoopla to market their estates. If online-agents displace high street brokers, then Rightmove / Zoopla will end up capturing all the economic rent anyways, so it seems hard to imagine an outcome with a redfin-type player in the U.K.

It is hard to imagine a business like Redfin can be competitive in the U.K., when for ~2% you can get represented by the Country’s largest real estate brokers. However, there surely are sites that are attempting to do so (eMoov, for example), and one must be cautious of this overarching risk[28].

Results So Far

In the first half of 2014, total income was up 29% YoY and EBITDA was up 70%. While the implementation of April’s MMR may result in a weaker 2^nd half than seasonally suggested, historically roughly ~1/3 of EBITDA is generated in the first half. If this pattern holds up, 2014E EBITDA could be in the neighborhood of ~£130 million (Bloomberg estimates are for ~£125 million[29]).

Here are some summary disclosures based on the end of the first half. The optimism might have been subdued somewhat in the last month or so with Foxton’s announcement (though we must remember Foxton is largely London based, which is at the forefront of the bubble mania):

• real estate pipeline up 27% at end of 1H vs last year
• Prime Central London prices cooled towards the end of the 1H
• Pipeline of lettings acquisitions for the remainder of 2014
• Total UK mortgage lending estimated to be around 200-210 million (15% increase YoY, in line with estimated tx volumes)
• Will probably beat peak EBITDA margins this year[30]
• Surveying services, excluding recent investments, would’ve been 1.4 million higher (“we haven’t generated a brass bean of revenue. You cannot let them go and value property until they get RICS certified.”
• 96% of mortgages will not involve Help to Buy

Risks:

• New Normal: The number of transactions in the UK may simply be lower here on out due to a permanently more “stringent” mortgage market (a moderation of the business cycle) and lower affordability for both first time buyers and second step movers
• Technology: As technology gets more efficient, the “middle-man” may be cut off, particularly in the lower service segment of lettings
• Exposure to High-End: Countrywide is over-weighted in the high end of homes (listed ~10% of homes >£500,000). Given that the average commission on a London&Premier home is ~6x that of a Real Estate home, a large decline in the high end markets, even while overall markets are recovering, could put a large dampener on earnings growth
• Discount Brokers: While figures suggest that sellers in the UK (and in the US) still see substantial value in hiring an expert to sell their homes, the existence of “low end” brokers that charge ~£500 instead of ~£3,000 to post homes on Rightmove and Zoopla could potentially disrupt the market and change the value that people perceive in hiring a high-street broker
• Advertising Portals: Rightmove and Zoopla are value-add real estate portals that command the majority of advertising dollars in the real estate market. They are an effective duopoly and have historically raised rates at mid-high teens %. If their dominance persists and strengthens, then they may be able to capture more economic rent through increased ARPA rates
• Acquisitions: Acquisitions have been done at very accretive valuations. There is no assurance that such opportunities will exist in the future
• Regulation: The government is attempting to curb a housing bubble and may put in more “caps” like it did with the April MMR (putting limit at 4.5x net income for mortgages) and the October limit on amount of mortgages that can be written at 4.5x net income (no more than 15%) – interest rate stress tests
• Professional Indemnity Claims: Have been a large detractor from earnings in the past few years. This is somewhat of a black box despite management stating that the statute of limitations will curtail claims here on out.

Valuation

• Base PT of £8.50 is based on ~14x 2018E EPS, or 10x EBITDA and ~18% IRR
  • Ø I assume transaction volumes return to 1.4 million by 2018, commission fees fall from ~1.75% to 1.5%, while average prices post 2014 increase at 2% (vs. OBR and Savills projections at 3.5%-4%). For high end homes, I model 1%-2% growth, and commissions falling from 2.0% to 1.8%. I model the Company’s incremental margins at 35%, falling to 30% (vs. 40%-60% guidance). I further assume the Company keeps leverage at 1x net, continues to pay a 2-2.5% dividend, spends ~40% of FCF on acquisitions at 7.5x operating profit (vs. 4-5x guidance), and the remainder on share repurchases
• Downside PT of £5.50 is based on ~12x 2018E EPS, or 8.5x EBITDA and ~7% IRR
  • Ø On the downside, transaction volumes fall to 1.1 million, average house prices stay flat, and commissions decline slightly to 1.70%. For high end markets, I assume -2% decline in both tx and pricing and flat commission rates. This leads to top line and organic EBIT decline of ~1% a year. I keep FCF uses the same

[model outputs]

Catalyst