Office Leases Lease terms very stable Trend toward short leases

Germany Food for thought Q1 2014 MARCH 2014 Office Leases Lease terms very stable Trend toward short leases ANDREAS TRUMPP Head of Research Germany Dr. TOBIAS DICHTL Research Analyst Frankfurt In market reports we usually refer to key figures such as take-up, vacancy and rental trends to help us assess current office market developments. However, the terms of signed leases are also essential when it comes to predicting medium and longterm trends on office markets. This information shows us how far ahead office tenants are planning, giving owners and investors an idea of average tenant turn o- ver. This study analyzes the terms of around 5,500 new leases signed in Berlin, Düsseldorf, Frankfurt, Munich and Stuttgart between 2008 and 2013. Location comparisons in all cities were conducted using commercial rental space (MF-G) as defined by the Society of Property Researchers, Germany (gif e.v.). Our assessment was based on the total lease term. We did not take early cancelation or renewal options or rent-free periods into account. Figures were initially calculated as unweighted and then followed by a weighted assessment based on rental space. This made it possible for us to evaluate the impact of large-scale leases. In our unweighted assessment, lease terms proved stable over time, for the most part, with a narrow corridor of between 5.1 and 5.5 years (see Ill. 1). In our asses s- ment weighted according to space size, the terms of medium-scale leases were somewhat RESEARCH CONTACT Germany Andreas Trumpp Head of Research andreas.trumpp@colliers.de Frankfurt Dr. Tobias Dichtl Research Analyst tobias.dichtl@colliers.de

Ill. 1: The development of the average duration of leases (in years) longer, with most large-scale space generating long leases and small-scale space being leased out for shorter periods. Based on this information, we identified a slight increase in the number of longer leases being signed starting in 2011 (see Ill. 1). Several very large-scale, long-term leases were signed in 2010 at a time when overall take-up was low, resulting in an aboveaverage annual value of 8.3 years. These included Vodafone in Düsseldorf with 76,500 square meters over a period of ten years, IngDiba and Allianz Global Investors in Frankfurt with 39,000 square meters over a period of ten years and 29,000 square meters over a period of almost eleven years, respectively. Ill. 2: Shares of office leases based on size of space During the period under review, around 64% of the leases we looked at were signed in the space segment of less than 500 square meters, with only around 3% being signed for space in the large-scale segment of more than 5,000 square meters. This indicates that the total space surveyed corresponds with the distribution of all registered leases in the top locations (see Ill. 2). Terms of leases relatively stable During the entire period under review, a solid two-thirds of leases were signed with a term of up to 5 years. The >3 to 5 year range was particularly popular with almost half of all leases being signed for this timeframe. In contrast, longer leases were considerably more rare. While the >10 to 15 year segment accounted for a 4% share, terms of more than 15 years were the exception with a 1% market share. Leases exceeding 15 years were only signed in a few cases by tenants such as state-run institutions, educational institutions and doctor s offices (see Ill. 3). 2 Food for thought Germany Q1 2014 Colliers International

Ill. 3: Leases based on term length Clear regional differences We saw the longest leases being signed in Stuttgart and Munich during the period under review, with unweighted averages of 5.5 and 6 years, respectively. This can partially be attributed to the considerably larger number of leases signed with a term of 7 years or more compared to the other top locations, each of which recorded around 19%. Results for the other cities were fairly well matched with average, unweighted leases of 4.9 years (Berlin), 5.0 years (Düsseldorf) and 5.2 years (Frankfurt). Leases of less than 3 years in Berlin generated by far the highest market share among the cities surveyed at a solid 31%. Combining these with leases with a term of up to five years increases the share to 80%. This may be attributable to the vibrant entrepreneur and start-up scene in Berlin as these companies are not yet certain of success and therefore tend to look for shorter leases. In contrast, large-scale leases were primarily signed by established, larger companies, which are to be found in fewer numbers (see Ill. 2). Short leases for existing buildings Differences in lease terms can be clearly seen among the different building types (see Ill. 4 and 5). It does not come as a surprise that the longest leases by far were signed for developments projects. We calculated an average term of 9.8 years unweighted and 10.8 years weighted based on size of office space. The gap between weighted and unweighted results was the smallest here at 1.0 years compared to the other building types, reflecting the significance of long-term leases on the one hand and large-scale leases on the other when it comes to pre-letting development projects. The second-longest leases following development projects were recorded for buildings under construction with 7.8 years unweighted and 9.3 years weighted. Leases for projects and buildings under construction typically comprised larger spaces. Among the leases under review, the median value for office space leased in projects was recorded at around 4,250 square meters with space leased in buildings under construction at 900 square meters. We saw leases with a term of more than 15 years being signed almost exclusively in projects, because these terms were typically included in the requirements for getting development projects underway. Ill. 4: Average lease term based on building type (in years) Leases for existing buildings and new buildings, both u n- weighted at 5.0 and 5.9 years, respectively, and weighted at 6.2 and 7.7 years, respectively, were considerably shorter than leases for projects and buildings under construction. Most of the leases signed for existing buildings were re-lets. In many cases, existing buildings that had not been renovated or only renovated to a limited extent and did not meet the latest tenant demands tended to and continue to attract more price-sensitive tenants who prefer shorter leases. Another factor here is the fact that leased office space in existing buildings is significantly smaller than that leased in new buildings since large-scale tenants tend to focus on more modern properties. The shorter leases we saw in new buildings compared to projects and buildings under construction can partially be attributed to the fact that the space being leased was often small or medium-scale first-occupation space in buildings no older than 3 Food for thought Germany Q1 2014 Colliers International

three years. We can assume that, in some cases, owners were more willing to compromise than with the first larger-scale rental spaces because they were more intent on fully letting the property. Ill. 6: Average lease term based on building quality (in years) Ill. 5: Share of leases signed based on property type and lease term (in %) Lease terms for renovated/refurbished buildings reflected a remarkable trend. With an unweighted average lease term of 6.0 years, these types of buildings came in third but still 1.0 year above the average for existing buildings. Weighted, they even caught up with buildings under construction. This reflects the benefits of extensively renovating existing buildings, since it becomes possible to lease space for longer periods of time and potential large-scale tenants once again become interested in the property. High building quality, long leases As expected, the longest average leases were signed for class A buildings, i.e., the most modern office properties. In our unweighted assessment, leases ran for around 1.0 year above average in class B buildings and around 1.5 years above average in class C buildings (see Ill. 6). The differences became more pronounced here as well in our weighted assessment based on space size because leases signed for class A buildings tended to be for more space (median of 498 square meters) than in class B and class C buildings (each with a median of around 300 square meters). Because the leases being signed for class A buildings were primarily for space in the development projects mentioned above, in buildings under constru c- tion and in extensively renovated buildings, the differences in rental space size and lease term are not that surprising. In contrast, class C buildings were often used as transitional space and were leased for shorter periods of time accordingly (see Ill. 6). Longer lease, higher rents? We observed an almost parallel relationship between rents and lease terms during the period under review. Lease terms with rents exceeding 20.00 per square meter were, on average in the assessment weighted based on size of the space leased, almost three years longer than leases signed at rents of up to 10.00 per square meter. Even in our unweighted assessment, they were still almost one-and-a-half years longer (see Ill. 7). This relationship, however, cannot be attributed to the assumption that longer leases mean higher rents. Instead, we are looking at an indirect relationship that can be explained based on building quality and building type. Ill. 7: Average lease term based on rent category (in years) 4 Food for thought Germany Q1 2014 Colliers International

High rents were paid in high-rises, prime locations and/or high-quality office buildings. Demand in this market segment is often met with a relatively limited supply of space, which improves the owners negotiating position. In addition, project developers and owners need to be able to plan on certain rental income over the long term and are also given guidelines by the banks providing financing regarding (large-scale) leasing and target rents. From the tenant perspective, many companies in exclusive locations and properties are willing to accept longterm leases and correspondingly higher rents to secure mo d- ern, centrally located office space. leases signed by doctor s offices, which tend to prefer longterm leases due to the high investment involved in moving into new office space. Doctors tend to stay in one place because of patient loyalty to their doctors at that specific location. Ill. 8: Average lease term based on location (in years) In the lower price segment, i.e., especially that of buildings of lower B quality or C properties, interested potential tenants have a considerably larger number of alternatives to choose from, forcing owners to accept shorter leases in this segment. Limited impact of location Our research showed that a building s location in the city only had a very limited impact on average lease term. While the unweighted average of lease terms in central locations in the cities surveyed ran for 5.5 years during the period under review, leases in other city locations ran an average of 5.2 years. This relationship did not change in our weighted assessment based on space size. In our weighted assessment, we saw average lease terms of 7.5 years in central locations and 7.3 years in other parts of the city (see Ill. 8). Although leases with terms of up to seven years tended to be more rare in central city locations, we did not see a difference significant enough to be reflected in the figures. We did notice that many of the planned and completed development areas outside central locations over the past five to ten years were considered attractive and generated comparable lease terms as a result. Differences based on sector When assessing the length of lease terms based on industry, we observed differences of almost two years in average, unweighted lease terms (see Ill. 9). We saw the longest leases in the health and social services sector at an average of 7.4 years, followed by public administration, associations and social institutions. Among the first of these, we particularly recorded The long leases signed by the public sector, associations and social institutions with an average of 6.3 years are not surprising, either, because these industries traditionally tend to sign long-term as well as large-scale leases. Examples include the lease signed by the German Federal Office for Building and Regional Planning in Berlin for 29,500 square meters (15 years) and the lease signed by the OFD regional tax office in Frankfurt for 14,500 square meters (30 years). Ill. 9: Average lease term based on industry (in years) List of abbreviations: HW = Health care and welfare, PA = Public administration, organizations, welfare agencies, INS = Insurance companies, EDU = Educational institutions, TT = Tourism and transportation, BF = Banking and Finance, MI = Manufacturing industry, RD = Research and development, CF = Consulting firms, BR = Building and real estate, RF = Retail and food service, IT = Information and telecommunications, OC = Other companies 5 Food for thought Germany Q1 2014 Colliers International

The shortest unweighted leases were recorded in the indu s- tries of construction and real estate, information and teleco m- munications and retail and gastronomy, each running 4.9 years. One possible explanation for this is the economic development in these industries. Due to the fact that the construction and real estate industries are highly dependent on the state of the economy, companies from these sectors tend to prefer shorter lease obligations. Companies in the retail and gastronomy sectors, which are heavily dependent on consumer trends, also tended to prefer an unweighted average of not quite 5 years. The comparably short leases signed by companies from the information and telecommunications industry during the period under review can most likely be (partially) attributed to the large number of start-up companies and their (at least for the time being) uncertainty of success. in this assessment, coming in third with an average of 8.0 years as the result of a number of small-scale leases, particularly by doctor s offices. Food for thought In our assessment weighted based on space size, the importance of long leases among certain industries became very clear. Companies from the tourism and transport sectors in particular, whose average benefited considerably from Deutsche Bahn AG s decision to lease 72,000 square meters for a period of 15 years in the Siberturm building in Frankfurt, and the public sector, associations and social institutions industries were clearly ahead of the other industries with leases going for 9.7 and 9.4 years, respectively. Companies from the information and telecommunications sector fell in the center of the pack with an average of 6.9 years thanks to some largescale, long-term leases in this segment (e.g., Vodafone for 76,500 square meters in Düsseldorf, Microsoft for 26,000 square meters and Google for 14,000 m² in Munich and Deutsche Telekom for 14,600 square meters in Frankfurt). The health and social services sector relinquished their top position Our survey shows that lease terms have remained comparatively stable over the past several years with no significant differences between central and peripheral locations. For owners and investors interested in status quo forecasts that means that investments for tenant fitouts will remain stable at the current level and that the number of tenant relocations can also be expected to remain constant, as can lease terms. The fact that, according to our figures, undertaking extensive renovation results in longer leases and higher rents may be of interest to owners of existing buildings. Our survey also shows that only one out of three leases for space in existing buildings is signed for longer than 5 years. Leases of more than five years are only feasible in existing buildings that are in excellent condition. That means that asking for more than five years could in many cases lower demand for the property and make it more difficult to lease the space. It is therefore essential to assess the chances of successfully landing a long-term lease before putting the property on the market. Tenants interested in large-scale, high-quality office space in modern buildings will have a hard time avoiding long-term leases since in many cases they will have to start leasing the space as early as during the project development design phase. Smaller companies are usually not able to lease in projects until the building is under construction, shortly before or after completion or once the first large-scale (pre-)leases have been signed. 6 Food for thought Germany Q1 2014 Colliers International

Ill. 10: Average unweighted lease terms in the top locations Our analysis showed a relationship, if relatively small, between economic trends and lease terms. There appears to be a delay of approximately 12 months here, which can be attributed to the fact that both supply and demand take time to adjust to changing conditions. Negotiations already underway are usually finalized based on the conditions specified at the start of construction. 5.2* 4.9 Glossary Building types 5.0 5.2 5.5 6.0 Based on the period under review, Hamburg 2008-2011 The terms of leases signed for refurbishments, that were equivalent to new buildings, reflect the benefits of extensively renovating existing buildings, since doing so draws the attention of large-scale tenants to these properties. Project: Projects refer to real estate development projects and office space before construction actually starts, i.e., generally until the foundation has been completed. Under construction: Buildings or office space under construction refers to the time between completing the foundation and handing over the keys to the finished building to the investor/owner. New building: New buildings refer to completed buildings or space that have been handed over to the owner and moved into but are not older than three years. Existing building: Existing buildings include buildings or space that were recently built but are older than three years or did not undergo extensive renovation before being re-let. Refurbishment: Buildings or space are referred to as renovated or refurbished if they have been extensively renovated, i.e., updated to a standard that meets current, modern tenant requirements, usually through extensive structural measures. Building quality Building quality is based on factors such as body/space set-up, facade, floor/electric, ceiling/lighting, heating and other fixtures and fittings such as building control technology and elevator systems. Buildings are rated according to these criteria as class A (very high quality, up to new building standards), class B (average quality) and class C (basic quality). 7 Food for thought Germany Q1 2014 Colliers International

482 offices in 62 countries on 6 continents United States: 140 Canada: 42 Latin America: 20 Asia Pacific: 195 EMEA: 85 $2 billion in annual revenue Authors: Andreas Trumpp Head of Research Germany +49 89 540411-040 andreas.trumpp@colliers.de Dr. Tobias Dichtl Research Analyst Frankfurt/Main +49 69 719192-29 tobias.dichtl@colliers.de Colliers International Deutschland Holding GmbH Dachauer Str. 65 D-80335 Munich +49 89 540411-050 1.12 billion square feet under management 13,500 professionals and staff About Colliers International Colliers International is a global leader in commercial real estate services, with over 13,500 professionals operating out of more than 482 offices in 62 countries. A subsidiary of FirstService Corporation, Colliers International delivers a full range of services to real estate users, owners and investors worldwide, including global corporate solutions, brokerage, property and asset management, hotel investment sales and consulting, valuation, consulting and appraisal services, mortgage banking and insightful research. The latest annual survey by the Lipsey Company ranked Colliers International as the second - most recognized commercial real estate firm in the world. colliers.com Copy right 2014 Colliers International. This document has been prepared by Colliers International f or adv ertising and general inf ormation only. Colliers International makes no guarantees, representations or warranties of any kind, expressed or implied, regarding the inf ormation including, but not limited to, warranties of content, accuracy and reliability. Any interested party should undertake their own inquiries as to the accuracy of the inf ormation. Colliers International excludes unequiv ocally all inf erred or implied terms, conditions and warranties arising out of this document and excludes all liability f or loss and damages arising there f rom. This publication is the copy righted property of Colliers International and/or its licensor(s). 2014. All rights reserv ed.