Introduction

One of the most important markets for PVC over the past thirty years has been the use of PVC for window profiles. At the start of the millennium this sector remains strong and is still growing across Europe. In other parts of the world there are also areas, such as China, where growth rates are astonishing. Specifically in the UK, the window profile market has given rise over the past twenty years to a parallel growth in foam profile. However, although volumes have been strong it is a different story when it comes to profitability in this sector and a number of issues need to be faced as the market in Western Europe approaches maturity.

Scope

The overall objective of the research programme summarised in this paper was to analyse the key technical and commercial elements in the window profile manufacturing chain, and to provide the industry with a solid basis on which they can make further decisions on investment and product development. The main part of this research has been carried out in the UK market but data has also been obtained to assess the position of Western European competition in this market.

European market breakdown

The proportion of PVC attributable to window profile has now grown to such an extent that it now accounts for 13% of the total PVC polymer market in Western Europe (Fig. 1). This sector consumed some 760Kt of polymer in 2000 (equivalent to 925 Kt compound). The German, UK and French markets account for the vast majority of this tonnage (approximately 85%) and the share taken by each country is shown in Fig.2. In addition there is a developing market in Central and Eastern Europe, and while usage is still relatively small, there is good growth. Currently, much of the requirement in these countries is supplied by imports, although local manufacture is relatively strong in some countries. The actual production amounts to around 50Kt of profile, with 80% of this being produced in Poland and Russia. In addition, Turkey stands out with 110Kt of local production mainly produced by four companies.

PVC-U is still the material most widely used for windows across Europe. In Germany, PVC-U is used for 55% of windows installed, while in the UK the proportion is about 45%. Overall in Central and Eastern Europe the proportion held by PVC-U is 38%. Recent statistics indicate that PVC-U holds 47% of the USA market while in China - a rapidly growing market for PVC-U - the proportion held by PVC is 30%, in a market which is about four times the size of the American market.

The nature of the European market means that there are considerable opportunities for the import and export of window profile as well as finished window frames - largely to the developing Central and Eastern European markets, with some 40% of profile being imported to these countries. Germany is easily the largest exporter of profile, while cost pressures result in the import of finished frames from adjacent countries such as Poland and the Czech Republic. In the UK the export of profile has remained fairly static at about 12Kt (5% of production) for a number of years but imports after falling by a third between 1994 and 1999 have shown a dramatic increase to a seven-year high in 2000 of over 35Kt of profile.

The technology for manufacturing window profile is essentially identical across Europe and the raw material companies - especially those involved in supplying the additives - as well as the machinery companies, operate global businesses in the sector. Since the developments made by these companies are available to all the major users, they operate in a highly competitive market and subsequently find that the critical issues they face are of a similar nature. There are about 70 companies manufacturing window profile in Western Europe, even after the significant mergers, and in some cases, elimination of companies which have taken place over the past four or five years. The market is thus still quite diverse compared with similar large applications for PVC, such as pipe or cable where the manufacturing capacity is concentrated in relatively few hands. However, the German and the UK markets in particular are being driven increasingly by cost pressures and are getting to the point where two or three companies will effectively dominate the sector.

In Germany and the UK there have been considerable changes in company structure over the past three years. For example, in 1998 there were 31 companies in the UK manufacturing window systems, while today there are just 21. Similarly in Germany, the number of systems groups has been reduced from about 26 companies to 16 groups. Thus the average output per group has increased in the UK from 7300Kt to 11500Kt and in Germany from about 19Kt to 27Kt. The mood for change which probably started in the UK in the mid to late 90's has spread to the German market and shows no sign of lessening, while competition in a fairly slow-growing market is increasing.

However, despite significant investment in technology, comparatively little investment has been put into the branding and marketing side of the business, which has resulted in fabricators discerning little or no difference between systems companies. In this situation systems companies are less able to support and maintain a premium price for their product. This, together with over-supply, leads to commodity-style purchasing, which then drives prices down even further.

There are substantial differences emerging at the beginning of 2002 between the structures of the German and the UK markets, as shown in Table 1 where it can be seen that there is a much greater concentration of market share in Germany, while in the UK there is still quite a long tail of companies with relatively small market shares.

Table 1 - Structure of business in Germany and the UK

Germany		UK
3 leading companies	% held by each company	3 leading companies	% held by each company
HTAG Group	29	Heywood Williams Group	12
Veka	16	Rehau	11
Rehau	10	Epwin Group	7
Total	55		30

We can see that the three leading companies in Germany hold 55% of production, while in the UK the leading three companies have just 30% of the business. In the UK no one company has a dominant share of the window market - out of the 12 largest grouping 10 have shares between 4.5 and 7%. In Germany we are close to reaching a position where one group has a commanding share of the market, but even there it is not certain that this will lead to critical mass being obtained.

A number of common technological elements have an impact on the profitability of the operation. These include: a) PVC-U window profile extrusion is a very large, competitive business and, therefore, efforts need to be concentrated on efficient and low cost profile production, b) the volume used and the number of systems offered by a profile manufacturer has a significant impact on efficiency and equipment utilisation and c) effective use needs to be made of the available capacity. Our investigations have been focused largely on quantifying and benchmarking the industry against these factors.

UK Window Market

The UK window market has been a mainstay of the overall growth in the PVC market for a number of years. While other applications such as pipe and cable have been at best stable, the window market has grown virtually continuously for many years, as shown in Fig.3. The growth data for the years 1994 to 2000 is shown in Table 2.

Table 2 - Growth in the UK manufacture of PVC-U window profile 1994-2000

	1994	1995	1996	1997	1998	1999	2000
Kt:	178	173	194	219	226	236	242
% change/yr:		-3	12	13	3	4	3

The market for PVC window profiles has increased by 40% in the last five years. The growth over recent years has been largely fuelled by the increase in the conservatory business and to a lesser extent the increasing use of PVC window profile for new build. As the growth in these markets settles back we forecast lower overall growth in the window market over the next few years. Thus from an average growth of 8% in the last five years we are forecasting a growth rate of around 2% for the next five years. The continuing developments in extrusion and ancillary machinery combined with improvements in tooling and calibration equipment have resulted in output rates steadily increasing. In the UK in 2000 there were 441 extruders being used for window and ancillary profile. If we examine the yield from each extruder we find, unsurprisingly, that the output achieved by the largest companies is significantly higher than those of medium and smaller sized companies. Of course, this is not the whole story and is just one of the benchmarks which can be used in assessing competitiveness and efficiency. Nevertheless, this yardstick does have a major bearing on capacity issues and also helps to identify the operations which may be running at lower unit costs.

These yield measurements can be used in different ways to compare operations. In the UK the average yield per extruder across the whole window business is 550 tonnes/ line, assuming all extruders are available and being used. This is a very marked increase from the situation in the UK a couple of years ago, when the average output per extruder was just 465 tonnes/extruder per year, and it shows the steps which have been taken by the UK window industry to improve their cost base. We have found that many companies also retain older extruders even when these are redundant to requirements. Ultimately it is demand that is the most important factor determining whether the available extruders are actually required. In practice, therefore, it is likely that the average yield is even higher than that stated above.

Average yields rise significantly with the volume produced, with the largest companies achieving average yields of 650 tonnes/machine and the smallest companies achieving yields of 175 tonnes/machine, as shown in Fig 4. However, the yield can vary widely even for similar sized operations and in the group of the largest profile manufacturers the yield range is 400 tonnes to 1150 tonnes per line.

Our research has also shown that after allowing for typical downtime required for maintenance and other factors common in the profile production sector, the industry is operating at just about 40% of available capacity. This level has not really changed much for many years and investment in new extruders continues at a steady pace. Over the years 1999 and 2000 there was a reduction in overall machines of five, while fifteen high-speed extruders were installed. Lean manufacturing methods practised widely in other parts of industry do not seem to have made an impact yet on the window profile sector, but when the leaders do turn to such methods they will gain significant competitive advantage. Whether any pause or downturn in the industry would result in a deeper evaluation of assets in the sector is yet to be tested as companies try to consolidate their position, but at a lower gross margin. Another aspect linked with capacity is the trend to faster tool changes apparent in some parts of the industry. While there is so much 'excess' capacity there is no imperative to move in this direction but some companies are certainly actively progressing this technology. This again will increase available capacity and will have yet another additive effect on production costs.

Many companies, even after mergers and acquisitions, are reluctant to rationalise the number of systems they manufacture. For example, in 1998 there were 88 systems available from the 31 producing companies. In 2000 there were still 88 systems but from a lower number of 23 systems groups. Our research shows that many companies do wish to reduce the number of systems but there are also conflicting pressures, such as the growth in sculpted systems and also pre-gasketted systems. The addition of some four and five chamber systems also adds to the total. The average tonnage produced per system in 2000 was 2750 tonnes but there was a big difference between larger companies and smaller companies, as shown in Table 3. The larger companies were able to manufacture over twice the volume per system than medium sized manufacturers and almost 3.5 times that of the smallest group.

Table 3 - Average tonnes per profile

Systems extruder ranked by volume	Output(Kt)	Associated number of systems	Average tonnage per system(tonnes)
Large (>10kt)	200	58	3448
Medium (1-10kt)	38	26	1460
Small (<1kt)	4	4	1000
Total	242	88	2750

We should not forget that a further 26 systems are imported into the UK, so the consumer certainly has plenty of choice.

Two companies were able to manufacture over 7Kt per system during 2000, while 16 systems companies - including some large groupings - were only able to manufacture about 3Kt per system.

In order to formulate their business plans the systems companies are naturally concerned about the details of the 'real' market for their profiles and not just in the volume produced by each company. Imports and exports of profile as well as issues such as stock build are important elements in the development of a supply/demand model for the business. Our research has examined this matter over a period of seven years and over this period we find that actual demand has risen by 42%, while the value of the market has increased by just 14%. As an example the added values for white and foiled profile are shown in Table 4.

Table 4 - Added values for window profile 1994-2000

Product	1994	1995	1996	1997	1998	1999	2000
Compound	100	100	100	100	100	100	100
Mainframe white profile - UK	219	185	213	209	233	221	204
Foiled mainframe - UK	370	324	382	369	370	369	366
White and foiled mainframe - UK	232	196	238	238	250	247	228

Sudden surges in the raw material price - which accounts for about 80% of production cost - cause major problems for the systems companies. Such a surge last occurred towards the end of 1999 and gave rise to companies desperately trying to protect their margins, the end result being that the weaker companies were progressively squeezed out of the market.

The importance of the conservatory market to the continuing growth of the window sector as a whole cannot be underestimated. Table 5 shows the large growth in the conservatory profile sector, which has underpinned the overall market over the past few years.

Table 5 - Annual growth rates for window profile and conservatories 1994-2000

Sector	% growth
	1994	1995	1996	1997	1998	1999	2000
Profile output volume: UK	-	-3	12	13	4	3	4
Profile sales volume: UK	-	-2	14	13	3	4	7
Conservatory installations: UK	-	8	10	11	9	10	13

UK Foam Profile Market

The foam market has followed the trend of the window profile market with a time lag of about ten years (Fig 5). During the period 1994 to 2000 the market increased by 78% with annualised growth rates of up to 15.5% (Table 6).

Table 6 - Foam compound usage 1994-2000

Year	1994	1995	1996	1997	1998	1999	2000
Volume (tonnes)	36700	42400	48500	55350	60450	65100	65500
% growth year on year	-	15.5	14.4	14.1	9.4	7.7	0.6

Although this rate appears to have slowed down over the past couple of years this market sector has been a star in the PVC business, and it is likely that 2001 has had particularly strong growth (well over 10%). In contrast to the window profile sector there are only 11 companies involved and the market volume in 2000 was 65.5 Kt. In this sector 137 extruders were used resulting in a machine yield of 480 tonnes/line/year. This can be compared with 413 tonnes/line only three years previously. The range of yields is very much smaller than it is for window profile at 410-500 tonnes/line - indicating that the technology and tooling is hardly differentiated across the market. The structure of the market is such that virtually all the companies involved in the foam market are also manufacturing window profiles. Again in contrast to the window market the only other country significantly involved in foam profile extrusion is Ireland, where a further 8.5 Kt were produced in 2000. In both of these markets there is a pattern of market dominance by two companies. In the UK the largest two processing companies hold 44% of the market while in Ireland this increases to 90%. Although there is some export of foam profile to Continental Europe this market remains substantially a UK and Irish business with two way trade existing between the two. We estimate that two thirds of Irish production was exported to the UK in 2000. The main applications for foam profile remain cladding, window associated trims and boards, and of increasing importance, roofline products.

In marketing terms the foam profile market offers a second bite of the cherry for window companies. Roofline products in particular are items which can be sold as an add-on to a good replacement window installation.

Many companies now see roofline as an alternative way to replace income, which is reducing from window installations. Quotes from numerous direct selling representatives suggest that roofline will be like PVC windows were in the early 90's. An essential difference between windows and foam profiles is that the foam profile range made by each company is virtually identical in shape, colour and performance. Thus brand differentiation is difficult to achieve. Some though are managing it.

While there was apparently a blip in the continuing growth in this market in 2000, forecasts suggest that short term growth of between 5 and 10% a year will be seen again in this market.

A good deal of effort has been put into taking cost out of the production operation. Nevertheless margins have been under constant pressure and there has been steady price erosion. Continuing improvements in technology - both materials and process - have helped and improvements in value engineering - such as reduction of foam density - have also helped keep costs down but in practice limits to how far this can go are being reached. Over the past five years a supply/demand model has been regularly prepared and this shows that while demand in volume terms rose by 41% over this period, the value of the market rose by just 6%.

An interesting offshoot of the growth in the sale of roofline products has been a significant increase in the use of PVC for rainwater goods. This very mature and largely stable market from the 70's to the end of the 90's has been very profitable for the PVC pipe and fittings manufacturers, who were the main players in this sector until recently. However, the replacement of fascias and soffitts using foam PVC profile has brought with it a parallel replacement of PVC rainwater goods and the volume of PVC used in this sector has now reached over 40 Kt for the UK and Irish markets.

Of equal significance, however, is that this surge in the use of PVC has seen a shift in the distribution network from a predominant supply situation using builders merchants and sheds to an increasing use of plastics distributors - who are already widely involved in the sale of foam profile as well as plastic sheet products.

Environmental issues

PVC has dominated the window and foam markets in the UK, Germany and France for many years. During the whole of this time there have been continuing environmental pressure against the use of PVC itself and in the past few years pressure was seen against the use of certain stabilisers such as cadmium and more recently lead containing systems. In most cases the PVC manufacturers and processors have successfully fought against these campaigns even though there have been occasional bans implemented on the use of PVC for windows. The last three or four years have also seen developments in alternative materials being promoted for use in windows such as polypropylene and even wood-filled plastics. These products have not proved their equivalence in performance to PVC and while some may be seen as a complementary product in certain specific situations they are not currently real threats to the dominance to PVC. Of current importance however, are recycling and lead replacement, both of which are bound to have some effect on the future of the overall sector and affect the overall economics of the business.

Recycling

The recycling of fabrication waste has now become an established part of the business in Europe - we estimate that at least 15Kt of fabrication waste was used in the UK window business in 2000. The EPPA (European Plastic Profile Association) has formulated a voluntary commitment to the EU where the industry will recycle 25% of the available collectible waste by 2003 and 50% by 2005. As part of this commitment the UK has recently agreed to evaluate a scheme whereby 2000 tonnes of post-use PVC is recycled in 2003, increasing to 4000 tonnes by 2005.

Since 1980 the amount of PVC profile produced for windows in the UK is a little over 2.5 million tonnes (Table 7).

Table 7 - PVC window profile production UK 1980-2001

	1980-84	1985-89	1990-94	1994-99	2000-01(est.)	TOTAL
Volume(Kt)	67	350	695	1079	500	2691

Our research has so far shown that PVC accounts for 5% of windows which are removed from properties. When this is further split into how the windows were disposed of we find that about two thirds of the PVC windows which were collected were ostensibly recycled (Table 8).

Table 8 - Proportion of windows recycled or dumped by installers

	Recycled	Dumped
Timber	8%	92%
Aluminium	41%	59%
Steel	57%	43%
PVC-U	66%	34%

One of the interesting facts to emerge from this work was that the average age of the PVC windows being removed was 18 years - less than that of the other materials (wood and aluminium). There are a number of reasons for this which require deeper study, but we are now getting to the stage in the UK where large quantities of PVC windows are going to become 'available' - at least theoretically - for recycling. It is clear that a management structure will need to be developed to cope with the situation, and a scheme will need to be devised whereby such windows can be effectively and economically recycled.

The replacement of lead stabilisers over a period of time will also form part of a voluntary commitment by the relevant trade bodies to the EU. Some German companies are actively selling profiles which do not contain lead, and one company covered by our research claims to have made the switch. One cellular foam company is actively promoting non-lead stabilised product, and while there is currently a cost penalty it is inevitable that the move to alternative systems will continue, although the rate of change is still unclear.

The EU has still not produced any formal views on the future of PVC and its applications, or the voluntary commitments made by the industry, although such views are expected to be given during 2002. However both of these important issues are crucial to the continued acceptance of PVC window and foam profile. For the industry as a whole it is probably a question of not 'if' but 'when' and it will need to adjust its business plan to prepare itself for the new requirements when they are introduced.

Future developments

There are a number of parameters which are critical to the growth of the window market. These include building stock, home ownership levels, income, competing materials, technical developments and product preferences. These factors drive the market and lessons are being learnt in the UK which can apply to other countries.

Recycling will have an increasing impact on prime material usage, with increasing quantities from both fabrication waste, and in the future, post-use waste being used. Factors such as higher extrusion speeds and more sophisticated extrusion technology, such as efficient twin strand extrusion, will increasingly favour larger companies who are able to provide the investment required. The high growth rates of the past look unrealistic and the scope will narrow for further sales penetration. The stock of housing available for refurbishment is rapidly declining. The implications of confronting these harsher market conditions will require a whole range of responses including greater price flexibility, more efforts to control end user markets by stronger branding, contractual changes and perhaps forward integration into fabrication. We will probably see more effective use of existing assets, restructuring of assets, disposal of surplus stocks and yet further attacks on costs. The market faces a number of challenges. Examples of these are:

1. Conservatories - they have been a boon, a swelling new market to compensate for declining sales in replacement doors and windows. Does it matter where a frame ends up? However, this is a one-off market and once installed there will not be a replacement market to follow it. What happens then? More importantly, conservatories are not as profitable to fabricators and installers as replacement windows and doors. Now they are a large part of companies' sales, firms will have to find a way to make them as profitable, or find something else to fill the profit gap.

2. Concentration - the trend to form fewer, larger companies to achieve benefits of scale and market power seems inevitable as winners take over losers. But although the acquirer expects the marriage to produce a bigger, more effective combination, benefits can be elusive. Four years ago the top 12 extruders in the UK accounted for 68% of total profile volume. This has grown to 83%, yet none of them has critical mass with enough power to influence the market in their favour. With just over 20% share of total market, large national retailers are even further from market power. Trade 'super-fabricators' are more confident of getting there, yet good profits, productivity and performance are spread throughout the fabricator size range. In theory the lowest cost producer should win hands down. In practice scale doesn't always equate with lowest cost or efficiency. There is a growing band of outstanding fabricators, a manufacturing elite that owes its success to smarter thinking rather than heavy investment or scale.

3. Replacing the replacements - As more of the market is replaced what's left tends to be part-houses, older, more difficult or isolated properties, and reluctant, more demanding or poorer buyers.

   There has always been a small amount of replacing the replacements, but the real challenge is to renew the market by starting to replace the replacements en- masse. Attractive, decorative windows, colour, higher levels of security, and vertical sliders will come into their own. Will these be enough to persuade large numbers of homeowners to spend their hard-earned money on replacing adequately performing PVC-U windows with higher specification replacements?

   And how soon is it before PVC-U windows are replaced? Although the PVC-U frame can last 40 years or more, in practice PVC-U windows seem to be replaced much sooner. On the face of it that's good news. But those are early windows. Sealed units, hardware and the appearance of profile are immeasurably better now, and consumers may not replace newer frames at anything like the rate we need. Will more energy-efficient windows do the trick? Or will windows that would be replaced anyway simply be replaced with higher specification, higher cost windows? The jury is still out on this.

4. Document L and FENSA look set to rule our lives in 2002-2003. What began with the laudable aim of significantly reducing CO2 emissions by raising the energy performance of the windows we install has been coupled to a scheme that appears to have been cobbled together at the last minute. Rushed through, with more questions than answers at the time of writing - just a month from launch - there is a danger that the scheme could backfire. Professional firms will be burdened with additional administration and costs, while 'cowboy' firms will continue to break the rules and exploit a growing price gap. If history is a guide, higher cost low-E glass will end up as a freebie, and the winners will be glass manufacturers and consumers.

We have seen evidence that the German market is more concentrated than the UK market. In both markets, power and financial health are difficult to achieve until relative critical mass has been obtained. In a stalemate situation we see companies indulging in pricing battles to maintain or grow their share. The principle of superior force and relative advantage does not guarantee success, but full frontal battles between equals almost always end in tears all round.