International topics Current Issues August 23, 26 German retail property Opportunities despite oversupply Important asset class. There is almost 12 million m² of retail space in Germany. Its value is estimated at between EUR 16 to 2 bn. The sales generated on this space aggregate almost EUR 4 bn a year. Drop in space productivity. For about the last ten years retail space has been growing considerably faster than retail sales, eating into space productivity since the mid-199s. Retail rents on the decline. Falling space productivity is reflected in falling retail rents. Edge-of-town locations in particular have come under pressure in recent years. This year, rents are set to stabilise only in prime locations. In the short term the two special effects of the World Cup and the value added tax increase taking effect on January 1, 27 are acting as a stimulus. But next year s damper is a given. Authors Tobias Just +49 69 91-31876 tobias.just@db.com René Stahl Editor Hans-Joachim Frank Technical Assistant Sabine Kaiser Deutsche Bank Research Frankfurt am Main Germany Internet: www.dbresearch.com E-mail: marketing.dbr@db.com Fax: +49 69 91-31877 Managing Director Norbert Walter Demographics are also depressing retail property. Demographic trends mean that price-adjusted retail sales in west Germany are likely to increase by only about 1 percent p.a. up to the mid-century, while in east Germany maximum growth of a scant.5% a year is to be expected. Yet international investors are still lining up to buy German retail property. First, the German market is by no means oversupplied with all formats. There is potential for catch-up in large retail formats. Second, the interest gap between rental yields and financing rates is still favourable. Greater polarisation. The middle ground offers virtually no opportunities for growth. Impetus is coming from price-aggressive discounters at one end of the scale and properties offering added value at the other, e.g. entertainment, emotional experiences or a special ambiance. Expansion in space despite sluggish sales Sales in EUR 196 1965 197 1975 198 1985 199 1995 2 25 21 Former West Germany until 199, whole of Germany from 199 Turnover per inhabitant Turnover per m² 6 5 4 3 2 1 Sources: EHI, Federal Statistical Office, DB Research
Current Issues Introduction Heavy investment in German retail property, despite marginal advance in sales For some years, extensive German residential portfolios have been snapped up mainly by foreign investors triggering extensive media coverage and increasingly intense political debate meanwhile. But for some time now investors have no longer been focussing on German housing alone. Instead they are turning increasingly to logistics, office and retail properties. Jones Lang LaSalle says that in 25 roughly EUR 6.5 bn was channelled into retail property. Atisreal even estimates the volume of transactions at just under EUR 8.5 bn. In each case this represents about a third of total industrial property investment. Market watchers expect a marked increase in 26. 1 Given that the 118 million m² of retail space is valued at a total of EUR 16 to 2 bn, current transaction activity is therefore considerable. 2 What lies behind this enthusiasm for German retail property? According to Federal Statistical Office sales data, real retail turnover in 25 was not even 3% up on its 1994 level. Indeed, in four of the last ten years retail sales have fallen. The frequency of German retail insolvencies soared between 2 and 25 from 62 to 14 businesses per 1, retailers. Yet at the same time retail space was expanded by more than 2%. With more space encountering flat sales, one wonders what appeal this asset class has for investors. This paper sets out to highlight the perspectives for investments in German retail property. We begin with a description of the German retail market before going on to describe the common property formats. The study focuses on analysis of the major market trends, deriving recommendations for investors, sellers and municipal planners from this. A close look is taken at demographic trends and the different regional developments associated with them. Declining share of private consumption (nominal) 14 12 1 8 6 4 2 EUR bn % 92 94 96 98 2 4 Consumer spending priv. HH (left) Retail turnover (left) Retail share (right) 45 4 35 3 25 2 15 Sources: Fed. Stat. Office, EHI 1 The retail trade in Germany All told, the retail trade generated sales in 25 of just over EUR 5 bn, making it one of the most important sectors of the German economy. Retail in the narrower sense, i.e. stripping out motor vehicles, fuels and pharmacies, sold goods in 25 worth roughly EUR 39 bn. 3 That represents an increase of around 1% on the previous year. But it also means that once again nominal retail turnover rose more slowly than private consumer spending as a whole. Meanwhile the retail sector accounts for only about 3% of private consumer spending; at the beginning of the 199s its share was 1 percentage points higher. Consumers in Germany are evidently turning more to goods and services outside the retail range and shifting their demand accordingly. This is suggested by high saturation levels for necessities and high levels of supply with consumer durables. Sluggish latter-year economic development is not enough to explain this; the problem is not that people are 1 2 3 Jones Lang LaSalle (26). Capital markets newsletter 26 and Atisreal (26). Investment market report Germany 26. See EHI Retail Network (26). Handel aktuell. Struktur, Kennzahlen und Profile des deutschen und internationalen Handels, Issue 25-26. See also Just, T. (appearing shortly). Einzelhandelsimmobilienmärkte. In Schulte, K.W. [Editor]. Immobilienökonomie IV. Volkswirtschaftliche Grundlagen. Hauptverband des Deutschen Einzelhandels (26). www.hde.de. 2 August 23, 26
German retail property 5% of sales from necessities Shares in retail turnover, % 11.4 2.5 14.7 Food an beverages Household goods Textiles, clothing Furniture, furnishings Others 18.1 35.3 Source: Fed. Stat. Office 2 Traditional stockists market share shrinking Location choice is crucial consuming less, but that they are consuming relatively fewer retail products. Sales of food and beverages in 25 accounted for almost 35% of retail sales, followed by household goods and daily necessities at 18%. Textiles generate around 15% of classical retail sales, with another 1% each going on furniture, technology and other goods. Household goods, furniture and food and beverages in particular hold out the prospect of only very limited growth, given than many consumers have literally reached saturation levels in these segments. Heterogeneous supply of space All told, in 25 retail in the narrower sense occupied 118 million m² of space. Since the turn of the century an extra roughly 2 million m² is being added each year. In the years prior to this, the expansion in space was considerably stronger. The supply of space in the 199s grew by more than 3 million m² in total. However, much of this can be explained by pent-up demand in the east of Germany, where the area on offer climbed from six to 18 million m² in the space of ten years. But even in the west of Germany retail space in the 199s was stepped up much more sharply than in the years after 2. Total sales space is divided among very different operating formats, which in turn are linked to very different commercial properties. The full spectrum extends from small food retailers, some with less than 1 m² of sales space, to big shopping centres in which the sale of goods forms only one facet of the overall range. While traditional stockists are still to the fore, they have been losing market share for years. Currently the figure stands at barely 25%. Then come specialist retail markets with a market share of 22% and, at upwards of 1% each, multiple non-food retailers and food discounters and cash-and-carry stores. User expectations of the property End users and sellers are traditionally brought together in a retail property for the purpose of presenting and selling goods. Consequently, retail area design and the choice of property location are crucial to the assortment and hence to the business s success. In a simple scheme of things, consumption can be distinguished along two axes. The first axis separates daily necessities from consumer goods and is therefore concerned with the frequency and regularity of purchase. The second axis tracks the price sensitivity of consumption, i.e. how crucial the price is to the purchase decision. Purchases of necessities, a category consisting mainly of food and beverages, are habitualised. Although price is frequently an extremely important determinant of the choice of store and product, ultimately the overall transaction costs of the purchase are the key factor, particularly in the case of frequent transactions. The location must therefore be quick and easy to reach. Consequently, properties are situated either directly in a residential area or on convenient transport routes. Here, too, there are also niche products that react only marginally to changes in prices (tobacco products, organic goods). Information, product display and advice are of minor importance to retail suppliers. Discount markets and supermarkets are examples of this type of property. August 23, 26 3
Current Issues Different types of demand require specific properties Purchase decision Determinants Information needs Examples of goods Property characteristics Recurrent demand Habitualised, partly impulsive Price, verifiable quality features One-off or infrequent demand After thorough informationgathering and preparation Features, services, price, image Low, therefore hardly High, assistance can influence the any assistance decision Necessities such as food, hygiene articles Plain, purpose-built, "low-price" buildings close to the customer Durables such as cars, furniture and jewellery Properties with special architecture in prominent locations, integration of other types of use; some big, out-oftown formats Examples of properties/ operating format "Emotional shopping" Discounters, supermarkets, very occasionally family business (organic products) Rather seldom Downtown arcades, retail warehouse formats, urban entertainment centres Increasingly prominent Source: DB Research 3 Additional services important with consumer durables Durable and semi-durable goods, on the other hand, require more information. Customers do not make such purchase decisions often and have to put up comparatively large sums. So while the costs of finding the goods and reaching the store are not insignificant, they are certainly not the prime concern. Price does not lose its competitive function either; in fact, price competition is particularly intense in consumer electronics and furniture. However, there are more segments among durables than among supplies in which price is not the most important instrument. Extra services often tip the balance here (delivery services, financing, assembly services etc.). Furniture, jewellery and consumer electronics are the most important categories of durables; textiles, household goods and toys the major semi-durables. Here, the provision of extra services moves to the fore in the properties and individual stores. 4 August 23, 26
German retail property Characterisation of retail properties Delicatessen trade Purchase decision price-insensitive Boutique Department store Shopping centre General department store Necessities Corner shop Trad. stockist Consumer durables Cash & Carry store Supermarkets Factory outlet center Specialist retail market Discounters Purchase decision price-sensitive Increase in importance Hardly any change Decline in importance Source: DB Research 4 Three latter-year trends Society and its values are in a state of permanent flux. These changes are also reflected in consumer habits. Retailers are therefore always at pains to model this constant transformation in their product presentation and the goods they select. In recent years three trends have been pivotal: first, the polarisation of consumption; second, retailers declining space productivity; and third, the resurgence of inner-city areas. All three trends will continue to have an effect for some years to come. August 23, 26 5
Current Issues Polarisation of demand Shares of retail sales, % 1973 1981 1986 199 21 Cut-price goods and dealer brands Middle market segment Premium-quality products Source: Eggert, 26 1 9 8 7 6 5 4 3 2 1 Expansion in space despite sluggish sales Sales in EUR ' 196 197 198 199 2 21 Former West Germany until 199, whole of Germany from 199 Turnover per inhabitant Turnover per m² Sources: EHI, Federal Statistical Office, DB Research 6 5 4 3 2 1 5 6 Polarisation of consumption The middle ground is dead. For the retail trade this means that the impetus to growth comes chiefly from high-end products and cutprice goods. According to Eggert, retailers then have precisely four directions in which they can develop to escape the dying middle. 4 They can move up or down with their prices, but as a rule competitors will already have occupied that slot. The other option is to add services to their product range. These can consist either of making shopping itself more convenient or of upgrading the property to a world of entertainment. This situation stems from three material shifts in recent years. First, big retail chains are making efficient use of all means of cutting their costs. Global product procurement and close cooperation with manufacturers create a delivery chain that smaller rivals cannot emulate. So there are significant economies of scale with standard products. Second, many consumers are now far better informed than they used to be. They can evaluate price differences and are prepared to pay higher prices only for better quality. This is intensifying price competition. Third, there has been a marked reduction in mobility costs in recent years. These days, consumers are far less reliant on stores within walking distance, meaning that price-sensitive goods can often be offered more cheaply in edge-oftown locations. This has also speeded up concentration in the industry, a trend accentuated by the tendency towards individualisation in society as people feel less bound by consumption patterns. Instead, they are guided more by their personal preferences and the constraints imposed by their incomes and less by group-specific moulds. Most of these drivers will remain important moving forward. It is, of course, plausible that mobility will become more expensive; also, an ageing society may be less mobile than a young one, which should make inner-city locations more valuable (more on this later). But these two counter-trends are unlikely to be powerful enough to reverse the polarisation of consumption. Declining space productivity We have already mentioned the mismatch between the rising supply of space and flat retail turnover. By European standards people in Germany do not spend much in the retail sector. In 25 each person in Germany bought retail goods for about EUR 4,7. Given a comparatively high savings ratio and the low proportion of retail sales in consumer spend, retail turnover per head of the population in Germany is well below the values in many other European countries. In France around 15% more sales are generated per inhabitant, in Sweden roughly 2% and in the UK about a quarter more. 5 This reluctance to spend, coupled with the declining importance of retail goods, puts pressure enough on retailers, and consequently also on retail property. A further challenge stems from considerably faster latter-year growth in retail space than in turnover. At present, turnover of around EUR 3,3 is obtained per square metre of sales space, almost 2% less than in the mid-199s and roughly as much as 2 years ago. What is more, these are nominal figures, i.e. not 4 5 Eggert, U. (26). Wettbewerbliches Umfeld Konsumenten, Lieferanten, Konkurrenten. In Zentes, J. [Editor]. Handbuch Handel. pp. 23-47. Admittedly this international comparison must be interpreted with care, because statistical registration of the markets is not yet fully standardised. 6 August 23, 26
German retail property Scale economies force selection processes Local authorities financial plight as catalyst of space expansion adjusted for overall inflation. EHI Retail Network (26) puts the supply of space per head in Germany, at roughly 1.4 m², at the top end of the range in comparable European countries. The Netherlands averages 1.1 m², France just under 1 m² and the UK a scant.7 m² per inhabitant. 6 Reasons for low space productivity At first sight the constellation makes no sense, with expansion in retail space amid sluggish sales. Nonetheless, this does follow strict business management reasoning and is made possible by a dilemma in the local authorities business location policy. Economies of scale through assortment variety. Some of the most important marketing parameters in the retail trade correlate positively to the supply of space. First, more space implies greater choice an advantage in itself for many customers. Second, large adjacent retail areas reduce transaction costs for customers able to make many purchases under one roof. This convenience also adds value. Of course, it need not necessarily lead to big retail warehouses, merely to more adjacencies in retail formats. These may also consist of shopping centres with lots of smaller shops. Finally, bulk suppliers can more easily streamline their delivery chains and lower their costs per sales unit by taking advantage of scale economies. That enables them to charge lower sales prices. These benefits also apply in the consolidation process. Competitors are obliged to increase the space they offer in the hope of growing market share in a flat market environment. This places them on the horns of a dilemma, because if all suppliers pursue this strategy their space productivity simply falls across the board. The competition situation is by all means comparable to cut-throat competition, implying that space is often on the market for a very long time. 7 The inevitable consequence is selection processes on the supplier side. The local authorities prisoner s dilemma. The number of residents and their purchasing power determine sales trends in a region. Where there is more than one local authority within a specific catchment area these compete directly for construction work, retail companies and consumers. The financial plight of many municipalities thus acts as the catalyst for the increase in space, with the authorities hoping to corner a somewhat larger share of the retail market and hence higher tax revenues. In many cases, however, this can involve no more than redistribution between neighbouring sub-markets. 8 In so far the local authorities find themselves in a situation resembling the prisoner s dilemma 9 in that, without cooperation, all local 6 7 8 9 All the figures quoted in this section are from EHI Retail Network (26). The textbook case of cut-throat competition describes a market with constant supply and diminishing demand. Competitors are then prepared to cut their prices to just above their variable costs. Although the company makes a loss, this is less than if it were to exit the market. Of course, this state of affairs cannot continue permanently in the long run the overheads must be reduced. In the retail trade the market functions similarly: with demand flat, supply increases and thus narrows traders margins. See Deutsches Institut für Urbanistik (26). Warnung vor Fehlentscheidungen bei der Genehmigung von Einkaufscentern in Innenstädten, position paper. The prisoner s dilemma describes a situation in which two decision-makers have to take a decision independently of one another that affects both parties. The incentives are such that, without cooperation, both parties take a rational decision that subsequently puts them in a worse position than before. Cooperation is the only way of resolving the prisoner s dilemma. August 23, 26 7
Current Issues Modern out-of-town retail areas compete with downtown locations Reawareness of city centres is also the result of fewer permits for greenfield sites Increase in downtown rents in prime locations Prime rents for 125 cities, EUR/m² 3 25 2 15 1 5 199 1995 2 25 Lowest level Highest level Median Sources: BulwienGesa, DB Research 7 authorities have an incentive to continue expanding space in a bid not to fall behind their regional rivals. Cooperation at local authority level, with coordinated space planning, is the only way to end this situation. It is, of course, important to keep a sense of proportion because competition between regions is not a bad thing in itself. Consumers benefit from greater choice, falling prices and closer shopping facilities. Supply growth makes sense only if consumers incomes and shopping behaviour alter. This must then also be mirrored in the property. Modernisation of the space on offer would therefore be essential and it is precisely here that the opportunities lie, because not all German retail formats bear the hallmarks of capacity overhangs. Reawareness of city centres The third trend is partly the result of intensified competition for retail space. The low-priced and often modern retail areas in big out-oftown shopping centres compete directly with downtown locations. Although not all lines of goods are in competition, the assortment in shopping centres has risen appreciably. This will create reawareness for the city centres among neither investors nor consumers but arguably among local politicians and downtown retailers. That is precisely what is now happening. There are pilot projects for Business Improvement Districts (BID) in Germany. BIDs have their origin in the USA and Canada. Their aim is to form a joint lobby and to coordinate business activities. One goal is to raise the effectiveness of building alterations or marketing campaigns. The key characteristics distinguishing BIDs from existing voluntary initiatives are, first, that retailers in the neighbourhood may be charged compulsory dues to avoid the inherent freeriding problem and, second, that BIDs are generally limited in time to prevent their compulsory nature from solidifying. There is not yet sufficient empirical evidence to judge whether BIDs can achieve their objective. They will presumably have sustained success only if local politicians play along by not designating any further new areas for development. This is precisely what local authorities are increasingly doing. At least with shopping centres, far more malls are now being completed in city centres than on greenfield sites. The main reason for this rediscovery of the city centre is the reduction in planning permission for developments outside municipal areas. In many cases, however, new formats are being established not directly in downtown areas, but on the fringes of prime locations. Rivalry therefore still arises with the city centre but this time within municipal borders. Current market situation The major retail locations in Germany Almost 9% of the people in Germany live in cities, seven percentage points more than in Western Europe as a whole. Even if Germany does lack a really cosmopolitan city like London or New York, there are very big conurbations. A total of eleven agglomeration areas exist, each housing at least one million people. The German urban landscape can therefore be described as a tightlyknit and closely linked network of cities. The hubs of these agglomerations are of such crucial importance to the retail trade because they exert powerful centripetal forces. 8 August 23, 26
German retail property Tourist and commuter turnover significant A distinction must be made here between the absolute and relative importance of these centres. Of course Berlin, Hamburg and Munich, cities with more than a million inhabitants, generate the highest retail turnover in absolute terms, too. But it is also well worth considering the more than 8 midsized towns and cities in Germany with populations in excess of 1,. As a rule they, too, are the heart of an economic area, radiating way into the surrounding region. To depict this relative importance, we usually take the centrality ratio. This calculates retail sales in a city relative to its purchasing power. A value higher than one means that retailers sell more than the inhabitants purchasing power might lead us to expect. Additional turnover from tourists or commuters plays a prominent part here. The cities with centrality ratios currently higher than two are Passau, Trier, Giessen und Rosenheim. Munich scores a mere 1.14, Frankfurt/Main a rating of 1.1. Besides the absolute size of the market, its relative importance should therefore always be analysed as well. The major German retail cities 25 Inhabitants (24) GfK POS turnover 1) Prime rent (1a location) Number Share of total Absolute Share of German retail sales ' % EUR m % EUR/m² Berlin 3,387.8 4.1 16,871 4.2 165 Hamburg 1,734.8 2.1 1,62 2.6 185 Munich 1,249.2 1.5 9,452 2.4 24 Cologne 969.7 1.2 6,572 1.6 18 Frankfurt/M 646.9.8 3,93 1. 2 Stuttgart 59.7.7 4,45 1. 2 Dortmund 588.7.7 3,59.8 15 Essen 588.1.7 3,425.9 14 Düsseldorf 572.7.7 4,232 1.1 175 Bremen 545.9.7 3,27.8 93 1) POS Point of Sale: altogether EUR 42 bn; roughly 3% higher than retail turnover in the narrower sense Prime rents more stable in cities Standard deviation of the rate of change from downtown rents, 1991-25 2 4 6 8 Number of inhabitants in ', 24 Cities with populations of over a million are not depicted to make the chart easier to read. They would not alter the result. West Germany 25 2 15 1 5 East Germany Sources: Fed. Stat. Office, BulwienGesa, DB Research 9 Sources: Fed. Stat. Office, GfK, DB Real Estate Relative and absolute stability of large markets It is precisely because retail property is often deemed a safe investment haven that it seems sensible to distinguish this perceived security by types of location. While it is correct that the standard deviation from prime rents in cities tends to be greater than in towns, this is due primarily to a statistical effect, because as a rule the rents agreed in cities are higher than in towns. The reason for this is that there is practically no scope for increasing the supply of inner-city space. As a city expands outward, so the catchment area widens and downtown locations become more valuable. The standard deviation in the rates of change is then the more accurate risk measure, since it eliminates the level effect. But standard deviation in the rates of change in prime rents in cities tends to be lower than in towns. Big cities are therefore relatively stable. This applies to cities in both the east and west of Germany. 8 August 23, 26 9
Current Issues Additional supply in Oberhausen and Mühlheim Turnover ratio, Germany = 1 Neighbourhood locations harder hit by lull in consumption Relation of prime downtown rents to prime neighbourhood rents 9 93 96 99 2 5 West Germany East Germany Average for roughly 1 west German and 25 east German cities Sources: BulwienGesa, DB Research 15 14 13 12 11 1 CentrO Rhein-Ruhr-Z. 9 Oberhausen Mühlheim/Ruhr 8 199 1995 2 25 Mühlheim/Ruhr Oberhausen Sources: GfK, DB Research 1 Marked rise in ratio of downtown to neighbourhood rents 6 5 4 3 2 11 Moreover, individual large-format properties in small and mediumsized towns can evidently cause greater market disturbances than in cities. The successful positioning of a new shopping centre can increase a town s centrality abruptly and significantly, but the repercussions on neighbouring towns are frequently negative. Even within a town, far-reaching adjustments can be set in train. The CentrO in Oberhausen highlights the opportunities for a location as well as the risks to neighbouring locations. When the shopping centre opened in 1996 with roughly 83, m² of retail space, it drove the centrality ratio up sharply, i.e. turnover in Oberhausen rose much faster than purchasing power. At the same time, however, the turnover ratio in neighbouring Mühlheim/Ruhr fell. The new competition was felt not only by neighbouring towns, though. Hardest hit was the centre of Oberhausen, as purchasing power was sucked out of downtown outlets into the new mall. Particularly in densely populated urban agglomerations with overlapping catchment areas there is a danger of such large-scale projects triggering a beggar-thy-neighbour spiral. In Mühlheim an der Ruhr the Rhein-Ruhr-Zentrum was developed into the Urban Entertainment Center in the post-1998 period. Since then, turnover has picked up again in Mühlheim while in Oberhausen the turnover ratio has plummeted. Initially, of course, this locational competition is a good thing for consumers, with both destinations at greater pains to remain attractive. It gives rise not only to additional, but also to modern and needs-based centres. Sustained drop in rents Higher retail rents can be obtained only if the location becomes more sought-after. As a rule this happens only when supply at a location increases more slowly than demand. Given that the German retail trade has posted only marginal sales growth in recent years while at the same time total space has been expanded and many retailers margins have remained under pressure, higher retail rents have been rare in the last few years. Of 1 west German cities, downtown rents in 85 are currently lower than ten years ago. On average 15% less rent per square metre is now paid than in 1995. The declines in inner-city locations are even more marked in the east of Germany, where they average slightly more than 25%. In neighbourhood locations prices have fallen even more steeply over the past ten years than in the city centres. Retail rents in 8% of all west German cities have dropped since 1995 by at least 2%. In east Germany almost 9% of neighbourhood rents eroded by at least 2%. On average east German neighbourhood rents are now more than 4% below their 1995 level. Consequently the ratio of inner-city rents to rents in secondary locations has risen appreciably in most cities. This holds true of both east and west German cities. However, it follows from the above that the ratio between the rents has risen chiefly because downtown rents have fallen less sharply than rents in edge-of-town districts. 1 1 A striking exception to this development is the retail market in Oberhausen. In the last ten years downtown rents have tumbled much more sharply than rents in neighbourhood locations. 1 August 23, 26
German retail property Rent increases in very few cities* Change in retail rents by location and direction of change, cumulated from 1995 to 25 West East City centre % % - Rising rents 13.9 12. - Decline max. 2% 44.5 24. - Decline more than 2% 41.6 64. 1. 1. Neighbourhood locations - Rising rents 6.9. - Decline max. 2% 14.9 12. - Decline more than 2% 78.2 88. 1. 1. *11 west German and 25 east German cities Sources: BulwienGesa, DB Research 12 Higher yield in east Germany reflects higher risk there Rents bottoming out Do initial yields send out the right signals? The net initial yields on west German retail property in city centres have risen over the past ten years by 3 basis points to roughly 6.5% (average of altogether 72 cities). Yields in particularly scarce locations such as Munich currently return just under 5%, while in smaller locations up to 8% can be realised. In the east of Germany initial yields have climbed since 1995 by 8 basis points and are currently about one percentage point above the level in the west. Here, too, there are marked regional differences. In Dresden the initial yield, at close to 6% is well below the east German level, in Görlitz, at almost 1%, clearly above. Both the trend and the yield differential between east and west German cities is plausible in the light of the market developments described. They clearly reflect the higher risk in the weaker retail markets, some of which post high vacancy rates. Even so, it is worth taking a closer look, and for this we have designed a simple indicator. We consider the prime rent at a location in relation to regional gross value added per head (in EUR 1,), to depict whether rents suitably mirror economic performance. We can then aggregate the values at city level into overall east and west German indicators. It is striking that in the early 199s the east German indicator was clearly above the west German level. Relatively speaking, east German rents were thus too high and had to come down. Meanwhile, though, they are bottoming out. At present rents in east Germany, measured in units of gross value added, are even cheaper than in the west. Of course, the potential for catch-up that this implies should not be overemphasised, as the indicator takes the supply side into account only indirectly. August 23, 26 11
Current Issues Rising yields = higher risk? Average values for 72 west German and 24 east German cities 6 4 2 8 7 6 91 92 93 94 95 96 97 98 99 1 2 3 4 5 Prime rent east (GVA per head EUR '), left Prime rent west (GVA per head EUR '), left Net initial yield east, downtown, in %, right Net initial yield west, downtown, in %, right Sources: BulwienGesa, DB Research 5 13 Advance of shopping centres 12 1 8 6 4 2 FI DK BE PT DE NL SE FR IT ES PL NO GB 65 7 75 8 85 9 95 5 Total area in million m² (l.) Number of centres (r.) 36 3 24 18 12 6 Source: EHI 14 Spreads still positive Rental yield less government bond yield, percentage points, shopping centres, 25 1 2 3 4 Sources: RREEF, CBRE, JLL 15 Foreign investor interest As with foreign financial investors euphoric attitude to German residential portfolios, some market watchers are also surprised at the pick-up in cross-border investment in German retail property, given the difficult retail environment in Germany. Some of the investment motives are certainly related. Ultimately, three aspects are involved: 1. Potential for catch-up in investment formats. Investors are interested not in the entire spectrum of properties, but only in the investment potential, or the properties already traded on the investment market. These are mainly shopping centres and other retail warehouse formats. Precisely in this segment Germany still appears undersupplied at least by the standards on other west European retail markets. Certainly, Germany does already have 36 shopping centres featuring total retail space of more than 11 million m², which is to say that the space on offer in shopping centres has almost doubled over the past 1 years. In 26 another 2 centres are expected to go onstream, and in the post- 26 period another 27 malls are in the pipeline. But this still means only 14 m² of shopping centre space for every 1, inhabitants, contrasting with more than 2 m² in Spain and France, while the Netherlands offer just under and Sweden well over 3 m² per 1, residents. This segment therefore tends to be in short supply in Germany. 11 Retail warehouse formats in particular are still less widespread than in other countries. That local authorities are shying away from substantial expansion of the property on offer in this segment to shield their downtown locations represents an opportunity for existing properties, since it keeps down the supply risk. 12 In the past few years even shopping centre developments have been mainly in downtown locations. Only about 4% of all shopping centres built since 22 are on greenfield sites, 11 The same can be said of the entire investment grade retail property segment. See O Roarty, B. (26). European retail economic prospects. Consumption, sales and prospects. Address for Experian Property Futures Conference. Emerging opportunities in European retail. London, April 26. 12 For further details see Just, T. and O Roarty, B. (25). European retail warehousing prospects. Interpreting diversity and drivers. Deutsche Bank Research. Current Issues. October 6, 25. 12 August 23, 26
German retail property Listless sales growth Retail turnover, % yoy -5 1951 1962 1973 1984 1995 26 Original values Hodrick-Prescott filter The Hodrick-Prescott filter is an algorithm that smoothes out short-term fluctuations in a time series. The resuls are often called trend components. Sources: Fed. Stat. Office, DB Research 25 2 15 1 5 16 compared to one in four up to 1995 (see the above section Reawareness of city centres). 2. Yield gap. Financial investors are also attracted by the stillfavourable interest rate environment. The comparatively high property yields contrast with still-low financing rates, creating an attractive effect for highly leveraged investors. The interest rate spread between rental yields and long-term financing rates is high in comparison to other European countries. 13 Although the gap in Germany is by no means the highest in Europe, all the markets with a higher spread are small and therefore less liquid. It is highly likely that sustained investor interest on the one hand and rising long-term interest rates on the other will narrow the interest rate gap in the further course of the year. 3. Chances of economic revival. Hopes of potential for catch-up are consequently mooted, statistically underpinned by the mean reversion concept. Or to put it more simply, a market will not be able to deviate permanently from its trend growth rate. But we need to keep a sense of proportion here. Trend growth rates in the German retail sector have been in decline for decades. It is true that trend growth has been rising again since 1995. But the increase is still very fragile and could be somewhat overstated by the 24 alteration in statistical collection methods. Short and medium-range outlook 14 Special effects as short-term drivers According to the official statistics from the Federal Statistical Office, the German retail sector posted real growth of about 2% in 24 and upwards of 1% in 25. In 24 the Statistical Office brought the group of reporting companies up to date, weighting younger companies more heavily. This explains the comparatively high rate of growth in 24. Another two special effects are set to add momentum in 26. The FIFA World Cup in the summer of 26 drew many football fans to Germany, who naturally also consumed there. What is more, regulations on trading hours were eased during the World Cup, slightly increasing the possibilities for consumption. Admittedly, neither effect will have a very great impact. Even if the extra million guests for the month of football fixtures consumed as much as the locals, this would only push up retail sales by.1%. Pull-forward effects more important than World Cup stimulus Arguably more significant for retail business than the World Cup will be the value added tax increase coming into force at the beginning of 27. Many people will want to avoid the extra three percentage points of VAT on consumer durables in particular and bring their purchases forward to 26. This effect will act as a stimulus in the current year, with a corresponding negative correction programmed for 27. German retail sales in the first quarter of 26 were up 1.2% in real terms on the same quarter of the previous year. With strong pull- 13 The interest rate spread can be calculated either as the difference between rental yields and government bond yields, which rather tends to reflect the relative attractiveness of the property investment over the safe alternative investment, or as the gap between the rental yield and the swap rate as a measure of the financing terms. Both spreads are positively correlated. 14 For more details see Auer, J. (appearing shortly). Perspektiven des Einzelhandels sind limitiert. Deutsche Bank Research. Aktuelle Themen. August 23, 26 13
Current Issues Competition from new distribution channels B2C sales in EUR bn 1.3 2.5 5. 8. 16.3 13. 14.5 11. 99 1 2 3 4 5 6 2 16 12 Risk diversification is always important 8 4 Source: HDE 17 forward effects in the second six months, a 2% increase is realistic for the year as a whole. But for 27 nothing but stagnation can be expected. Medium-range trends These economic ups and downs aside, four more trends will impact Germany s retail property markets in the medium term. Added competition from new distribution channels. New technologies are reducing the necessity for face-to-face contact between buyer and seller in a salesroom for the act of purchasing. The swifter data transmission works, the better goods can be presented online. The rising number of internet users and increasing technological convergence are further raising the potential for online transactions. 15 Growth rates in online sales to end-customers (B2C sales) in recent years have been in the double digits although admittedly from a very low base level. Rates well into the doublefigure range are to be expected for the coming years in this segment. Even so, the internet is not sounding the death knell for the stationary retail trade. For one thing, internet trading often gains at the expense of conventional mail order business. Second, growth in B2C in recent years stems less from the sale of goods than services. 16 Third, additional opportunities are even emerging for retailers using the internet as a platform to advertise their stationary range to customers. The substitution potential is greater with standardised, non-perishable goods than with perishables or goods in need of explanation. It is estimated that at the end of 215 online trading will still only account for around 1% of retail trading. 1. Increasingly international competition. Consolidation of the real estate sector will persist. With free capital markets, this will continue to take place internationally. National particularities will become less important as consumption patterns are internationalised and companies undergo a rapid learning process. At the same time the retail property market will grow more professional and also more global. Both these trends follow two fundamental laws of economics. The first is the law of the division of labour. Like office users, retailers will realise that it is often not necessary to carry the property they use on their own books. Second, the law of risk reduction through diversification also applies to property portfolios. As regional markets develop differently, more efficient and internationally mixed portfolios can be composed. 2. Waning importance of retail suppliers. German retail suppliers operate in an intensely competitive environment. This places a permanent damper on price inflation. With saturation levels reached (Engel s law) and inflation low, retail suppliers relative importance is diminishing. Initially this applies only in relation to consumption as a whole. Although retail suppliers will still be very important, impetus must come from adding value to the sale of goods. The focus will then increasingly be on entertainment, convenience and fun value so-called emotional shopping. Four of the ten principles for shopping centres listed in a recent 15 See Stobbe, A., Just, T. (26). The dawn of technological convergence, E-conomics 56. Deutsche Bank Research. May 26. 16 See Pago etransaction Services (26). Trends im Kauf- und Zahlverhalten in den relevanten E-Commerce-Branchen im Jahr 25. 14 August 23, 26
German retail property Trading hours should be relaxed further study by the Urban Land Institute can be assigned to this added value. 17 3. Longer trading hours. Trading hours have already lost much of their significance. People have become more mobile, in border regions they already shop abroad where regulations are more flexible, petrol stations have steadily broadened their range, traffic hubs such as airports and railway stations offer shopping almost round the clock and the internet is open 24/7. This de facto watering down of regulations makes further liberalisation of the stationary retail trade a sensible way of creating a level playing field for conventional retailers. Although significant sales growth is not to be expected from this, since households disposable incomes will not alter, it could staunch further losses in sales to new distribution channels. Implications of demographics Fewer children and more pensioners Three important demographic trends Demographic trends can be forecast relatively well, as they are based on tenacious, cumulative processes and cannot be reversed within a few years. This is especially true of natural population growth, given that the seniors of 25 were born, say, thirty years ago. Of course there are also unknowns in population forecasts: more or fewer immigrants than expected may come to Germany, we might live longer than presumed or we could have more children than the number on which calculations are based. However, since the birth rate in Germany has hovered at only just under 1.4 children per woman for more than 3 years, the potential number of mothers has dropped. This echo effect means that even a perceptibly higher birth rate would not reverse the tendency to a shrinking population. As a result the major trends can, at best, be checked, but not reversed. For retail property, three trends are pivotal here. First, the number of people living in Germany will decline by almost 1% by the middle of this century. Second, society is ageing fast. The proportion of over 65 year-olds will climb from about 17% to roughly 3%. Third, regional differences will become more pronounced moving forward as young, skilled people migrate to the dynamic economic regions. The Federal Office for Building and Regional Planning predicts that by 22 only 55 of the altogether 96 planning regions will see their populations increase. By 25 there will be only 35 regions in which more people live than they do today. In one in every nine regions the shrinkage will even top 3%. 18 17 These four points are: within malls department stores are losing their anchor position to public places and gathering spots (entertainment etc.); better networking of all parts of a mall; design parking as more than a ratio; deliver a sense of community (see Urban Land Institute (26). Ten Principles for Rethinking the Mall, Washington). 18 See Bundesamt für Bauwesen und Raumordnung (26). Raumordnungsprognose 22. Bonn. August 23, 26 15
Current Issues Consumer priorities are age-specific Consumption by age groups per household and month, in % > 8 7-8 65-7 55-65 45-55 35-45 25-35 < 25 Retail spending also agespecific Retail spending by age groups per household and month, in EUR > 8 7-8 65-7 55-65 45-55 35-45 25-35 < 25 % 5% 1% Food, drink, tobacco etc. Clothing and footware Living, energy etc. Interior decoration, household appliances Healthcare Transport Other goods and services Source: Fed. Stat. Office 18 1 2 3 Uses of relevance to retail trade Uses not relevant to retail trade These trends will act as a drag on the retail trade Demographic development is having a fourfold impact on the retail sector. 1. Number of inhabitants. It goes without saying that, all else being equal, fewer people will consume less. Population decline thus causes strains. 2. Population structure. Older people concentrate their spending on different things than young people. Two effects come into play here: life cycle effects, i.e. the age-dependence of consumption, and cohort effects, i.e. the dependence of consumption on the generation in question and hence on its particular formative stage. Together, the two effects result in considerably lower household incomes for seniors than for people in middle age, as revealed in the latest EVS sample survey of income and expenditure. This comes as no great surprise after all, in old age the most important source of income, wages and salaries, is no longer given. Consequently, consumer spending per household in old age is comparatively low. 19 What is remarkable, though, is the large difference in the patterns of consumption. For example, almost 4% of spending by the highly aged goes on living, almost 1 percentage points more than in younger households. Seniors also spend comparatively large sums on health and care. For that they economise on transportation and communications. It is interesting that the share of retail spending in total consumer expenditure holds relatively constant in all age groups. 2 The change in age structure would therefore impact chiefly on the retail product mix; but, if at all, it would have only a marginally negative impact on retail sales as a whole. 3. Disposable incomes. Business momentum in an economy is generated either by more economically active persons and consumers, more capital or technological progress. If the latter two factors are held constant, the demographic trends imply a direct reduction in trend growth. The incomes of a shrinking and greying population will increase more slowly than in the past. 21 4. Change in saving habits. Older people save less of their income than people in work. However, there is no evidence so far to suggest that older people use up their reserves by the end of their lives. In most cases their savings ratio is merely lower than before. But that still does not mean that the economic savings ratio will decline as the population ages, because particularly in the coming years economically active people will have to save more to compensate for falling pension payments. The net effect is thus by no means easy to calculate. Economic models suggest that the macroeconomic savings rate will edge lower in the coming decades due mainly to the pressure on interest rates. In Sources: Fed. Stat. Office, EVS 23, DB Research 19 19 Given that the average size of senior households is far lower than the average German household size, per capita analysis produces a more favourable result. 2 For these statements all categories of expenditure were classified by their significance for retail. As a further restriction, the share of retail turnover in total consumer spending by all age groups was set at 3% (see above). The highly aged had the lowest retail share, at just over 28%, followed by young people with 29%. The shares of the middle age cohorts ranged between 3 and in excess of 31%. 21 See Gräf, B. (23). German growth potential: facing the demographic challenge. Deutsche Bank Research. Current Issues. December 23. Frankfurt am Main. See also Gräf, B. and Schneider S. (24). Das Wachstumspotenzial kaum Chancen auf Verringerung des Ost-West-Abstands. Deutsche Bank Research. Aktuelle Themen 36. pp. 36-45. 16 August 23, 26
German retail property Long-range retail sales forecast 25=1 8 25 22 235 25 Migration winners and losers Population change in planning regions; in % Landshut Ingolstadt Bavarian Oberl. Se. Upper Bav. Lake Constance Munich Emsland HH-s. environ Upper Rhine-B. Göttingen Halle/Saale Altmark Upper Lausitz-N. Chemnitz/Ore.M. N. Thuringia Sw. Saxony East Thuringia Dessau -6-4 -2 2 22 vs. 25 25 vs. 25 18 16 14 12 1 W. Germany, end of structural change E. Germany, end of structural change W. Germany, ongoing structural change E. Germany, ongoing structural change Sources: Fed. Stat. Office, DB Research 2 Source: BBR 21 a greying society labour will be in shorter supply and wages will rise more steeply than the return on capital in which case the savings ratio will fall. 22 If we combine these four aspects, scenarios can be designed for the development in retail sales. Long-range retail sales forecast The following describes two simple scenarios, both based on the same macroeconomic environment. Essentially they follow earlier publications by DB Research (see footnotes 21 and 22). The average rate of growth in disposable incomes will slow in west Germany from around 1.7% p.a. to 1.1%. In the east of Germany it will range between.5 and 1%. But because of the gradual slide in the savings ratio (see above), private consumer spending will expand a little faster. In our first scenario ( End of structural change ) it is assumed that the share of retail in private consumer spending will not continue to fall, but hold steady at around 3%. In west Germany retail sales would then advance by 1.3% a year up to the middle of the century, despite the demographic pressures. Even in the east of Germany sales would expand by a good.5% p.a. And remember, these are real, i.e. inflation-adjusted figures. However, in view of the structural problems outlined above, with extra competition from new distribution channels and high saturation levels for retail goods, the assumption of constant retail shares is extremely optimistic. We therefore place a second scenario ( Ongoing structural change) alongside the first. In this scenario the share of retail in private consumption steadily loses another 1 percentage points over the coming 45 years to hit 2%. This will obviously dent the retail trade. In the west of Germany the average rate of growth will slump to a scant.5% p.a., while in the east retail sales (in real terms) would even show a steady decline; by the middle of the century they would be more than 1% below the present level. Three aspects are particularly important here. First, these forecasts must be interpreted with care as some assumptions have been made across regions and the economic forecasts narrowed down strongly to the demographic factor. This was done purely to simplify the analysis. Second, regions will have formative possibilities in future, too. Internal migration trends are not written in stone, they can be influenced (to a certain extent). At the national level, of course, this is a zero sum game. Third, in the above analysis a distinction was made only between east and west Germany. But even the level of the German states, on which the calculation was based, still appears far too coarsely layered. Catchment areas for retail properties are considerably smaller. The margin of fluctuation in population development in the individual planning regions is far higher than in the German states. The Federal Office for Building and Regional Planning predicts that the number of residents in Landshut and Ingolstadt will climb by well over 1% through 24, while in some east German planning regions population shrinkage will reach levels of more than 35%. 22 Vgl. Gräf, B.; Schattenberg M. (26). The demographic challenge: Simulations with an overlapping generations model. Deutsche Bank Research. Current Issues, May 19, 26. Frankfurt am Main. This would then also imply that capital does not increase at a constant rate as assumed above. Disposable incomes could even rise more slowly. August 23, 26 17
Current Issues Disposable incomes and disposable time as restrictions Emotional shopping on the advance Discounters still important In a growing number of regions the demographic burdens will outweigh the moderate productivity growth. As a result, retail sales in those areas will steadily dwindle. Without deconstruction, this will intensify the problem of space inefficiency described above. But even in most inward migration regions demographic trends narrow the growth perspective. In the longer run no more momentum is to be expected than in the past ten years. Properties for time rich and time poor Notwithstanding the generally anaemic long-range growth perspectives, demographic development also offers investors opportunities, inasmuch as the population trends and related economic effects will weaken middle retail formats and strengthen special formats. Not all properties are therefore equally affected by the demographic burden. Essentially, retail consumption is restricted by two factors, disposable incomes and disposable time. The time restriction refers, for example, to trading hours and to the accessibility of retail properties. The greying of society means that the group of comparatively timerich 23 will expand significantly. Within this group there will, on the one hand, be more wealthy pensioners benefiting from accumulated and bequeathed assets. On the other hand, previous decades of mass unemployment in Germany are likely to mean an increase in pensioners with low pension entitlements in future: poverty in old age is a real threat. Of course we do not expect the broad middle ground to be entirely eradicated between these extremes. Moving forward, the majority will still be located in the middle but growth momentum will be created at the edges. The two growth groups will shop in different retail formats. Wealthy seniors will seek the shopping experience, entertainment and diversion and will be able to afford to have all this close to where they live. For low-income seniors, on the other hand, discount formats will remain indispensable. Restricted senior mobility is something that discounters must allow for in their retail concepts. This applies particularly to greenfield formats that are difficult to reach. Not only will time prosperity increase, time poverty among another section of the population is also a logical consequence of demographic development. As the size of the labour force shrinks and financing burdens become increasingly onerous, longer working times are a given. International competition is piling on added pressure. The measures introduced by the government to raise the retirement age, and increases in weekly working hours as a new collective wage bargaining tool clearly reflect this development. Time is therefore getting scarcer for the working population. This makes better service, e.g. more home deliveries, and greater efficiency more valuable. But for people on low incomes there is no alternative to discount formats. Accessibility will become an even more crucial factor. While this does not necessarily mean being close to residential areas, it certainly does require good transport connections. 23 This term was used by Baring, A. (1997), for example, to denote people with a lot of free time at their disposal. These include seniors, and also the unemployed. See Baring, A. Scheitert Deutschland? DVA. 1997. 18 August 23, 26
German retail property Demographic trends additionally weakening the middle Property demand determined partly by constraint factors Time availability high low Financial security low high Target group Target group Wealthy seniors Highly qualified, mobile workers Property characteristics Property characteristics Entertainment value, near Efficiency, 24-hour service to home convenience The middle Target group Target group The impoverished age Low-skilled workers Property characteristics Property characteristics Discount formats, retail Discount formats, retail warehouses, only limited warehouses, tending to be greenfield location close to home, partly out of town Source: DB Research 22 Concluding remarks Yield spread depends on interest rates Net initial yield (NIY), in % 1 9 8 7 6 5 4 3 2 1 9 93 96 99 2 5 Government bonds (1-year) NIY retail warehouses west Germany NIY high street west Germany Inflation rate Sources: OECD, BulwienGesa 23 Retail property is often described as a safe haven because a large part of consumption goes on satisfying everyday necessities. In so far retail properties particularly in comparison to office property offer stable cash-flows. German retail property appeals to international investors in particular because of the upside potential they see in it. For one thing, there are by no means too many big shopping centres in Germany relative to comparative international statistics; if anything, there are too few. What is more, yields are still above financing rates. Finally, at least in the short to medium term, we have identified positive signs of a pick-up in retail business. Even so, investors must be aware of the specific risks and limits of the German retail property market. For years, retail sales have posted below-average growth in comparison to other countries and relative to private consumption as a whole. Many sales markets for typical retail goods are reaching the limits of growth, with new distribution channels via the internet and demographic trends in Germany further narrowing their scope. There is thus increasingly little justification for growth in space. However, this is just the broadbrush picture, because there is indeed upside potential for individual formats. But any new space coming onstream then reduces the chances for existing formats. As competition between conventional and modern retail space intensifies, the onus is on local politicians to develop sustainable retail strategies for their communities. That will be no easy task, since it is not just a matter of reconciling the interests of existing retailers and new players on the block. It is also about how best to accommodate consumer preferences. Grandfathering, which can lead to higher retail prices, does the shopper a disservice. Besides August 23, 26 19
Current Issues Sustained retail strategy by local authorities often requires cooperation between more than one authority which, new projects generally create new jobs in a community, at least temporarily, and the chance to suck purchasing power out of neighbouring areas. This latter aspect in particular shows that sustainable retail strategies require cooperation between local authorities. Isolated strategies often fail because they cannot resolve the prisoner s dilemma in which communities are trapped. Tobias Just (+49 69 91-31876, tobias.just@db.com) René Stahl Copyright 26. Deutsche Bank AG, DB Research, D-6262 Frankfurt am Main, Germany. All rights reserved. When quoting please cite Deutsche Bank Research. The above information does not constitute the provision of investment advice. Any views expressed reflect the current views of the author, which do not necessarily correspond to the opinions of Deutsche Bank AG or its affiliates. Opinions expressed may change without notice. Opinions expressed may differ from views set out in other documents, including research, published by Deutsche Bank. The above information is provided for informational purposes only and without any obligation, whether contractual or otherwise. No warranty or representation is made as to the correctness, completeness and accuracy of the information given or the assessments made. In the United States this information is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this information is approved and/or communicated by Deutsche Bank AG Frankfurt, authorised by Bundesanstalt für Finanzdienstleistungsaufsicht. In the United Kingdom this information is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange regulated by the Financial Services Authority for the conduct of investment business in the UK. This information is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this information is approved and/or distributed by Deutsche Securities Limited, Tokyo Branch. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Printed by: HST Offsetdruck Schadt & Tetzlaff GbR, Dieburg ISSN Print: 1612-314X / ISSN Internet and e-mail: 1612-3158