RISKS ASSESSMENT IN REAL ESTATE DEVELOPMENT : AN APPLICATION OF ANALYTIC NETWORK PROCESS

Sukulpat Khumpaisal and Zhen Chen School of Built Environment, Liverpool John Moores University Liverpool, United Kingdom
Abstract: A novel analysis approach to support real estate developers in decision-making to deal with potential risks in each project development stages is introduced in this paper. The analysis model applied in this research is the multi-criteria approach, based on Analytic Network Process (ANP) theory. Risks assessment criteria to accomplish the ANP calculation are defined based on both literature review and related experience, against Social, Technological, Economic, Environmental and Political (STEEP) requirements of the real estate development in order to assess the risks in this business effectively. Case study of a residential and commercial mixed-use project in Liverpool city centre is used to demonstrate the effectiveness of the ANP model. The result of an experimental case study result reveals that ANP is an effective tool to support developers in decision-making based on risks assessment. The ANP model is proved in its efficiency and can be adopted by real estate developers to suit for the business needs in order to assess risks in a real estate development projects.. Keywords:

Analytic Network Process (ANP), Real Estate Development, Risk Assessment, STEEP Factors

1. INTRODUCTION Risks and uncertainties are occurred in all real estate development projects, particularly in the complicated real estate project. Risks can strongly influence to each project stages from the project conceptual, project feasibility analysis, design and planning, bidding and tendering, construction and execution, and handover stage. Khallafalah (2002) states that the risk management process are generally an ongoing and iterative process, even each real estate project is different and unique. The typical approach of project risk management is three basic steps, which are; Risk Identification and Initial Assessment; Response and Mitigation and further Risk Analysis. Risks in this real estate development business are arisen by several factors as Social, Technological, Economic, Environmental and Political factors or STEEP as defined by Morrison (2007), Gehner, et al. (2006) and Clarke (1999). For example, risks in real estate development have been considered, in relation to the separation of design from construction, lack of integration, poor communication, uncertainty, changing environment and increasing project complexity and economic changes such as inflation and deflation, regional economic crises including greater competition in this business. Thus, risks and their consequence of risks caused by STEEP factors, must be considered and should not be underestimated, since those risks will impact overall project management processes, in regard to project programme delay, project cost overrun and the usage of the property, which cause a huge lost in project income. IPF (2007) defines that the nature of real estate projects are mostly the income generated properties, this paper therefore emphatically focus on the risks, which necessary concerned
1 2
Mphil/PhD Student, School of Built Environment, Liverpool John Moores University Senior Instructor, School of Built Environment, Liverpool John Moores University

when developers conduct the project feasibility analysis, because of feasibility analysis is a significant tool to assess the real estate project vitality in regard to forecast uncertainties towards Return on Investment in term of Income stream. The existing Risk Matrix method, is generally accepted by the real estate developers as the practical risk assessment tool to assess the consequence of risks. Kindinger (2002) and ioMosaic (2002). Younes, et al (2007) uses the investment of hotels to prove the capability of this Risk Matrix, it is also accepted particularly in the real estate investment. However, the data used for matrix calculation derived from either panel discussion or ranking method, which mostly rely on personal opinion rather than using quantitative measurements, and do not use reliable tools or instruments with strong theoretical basis. (Please see Figure 1 ANP and Existing Risk Assessment Model) Other inconvenience is that the risk matrix do not allow the comparison of each criteria, and results calculated by matrix are normally subjective, do not provide the detail of data to help the developers to structure their decision- making process. This is because risk factors are numerous, particularly in large real estate projects, and the ability of humans to assess many factors at the same time is very limited. He (1995) Additionally, according to the income generated characteristics of real estate, developers also assess the risks by measuring the investment return. Sagalyn (1990) suggest several financial ratio such as capitalization rate, Net Operating Income, Profit/Equity (P/E ratio) and discount rate to be raised in project feasibility as the tools to forecast the income generated and loss in real estate project. However, Wheaton, et. al (2001) and Hendershott (2002) argue that prior to construe these financial ratio, the developers mostly depend on the past and subjective information which are inaccurate to predict the future risks and the value of the newly established real estate project. According to the results of IPF survey in UK real estate industry, Frodsham (2007) reveal that real estate risks could be managed within an overall framework or risk management processes, those risks shall apply a variety of complimentary approaches, which are grounded in a rigorous and preferably quantitative framework. Therefore, the risk management processes shall include an assorted mix of Quantitative statistical framework as well as several techniques such as stress testing and a rigorous analysis of subjective issue. In order to asses risks and their consequences, it is suggested to use the practical tool, which could analyse risks, their consequences and computed the results in a numerical format. The desirable methodology for this real estate development should allow for the synthesis of the criterion, comparisons of each factors and to help the developers to structure the decision making process. Booth, et al. (2002), and thus, risk assessment process in the real estate development shall be supported by the modern method of mathematical statistics. Titarenko (1997) This paper, aims to introduce an application of ANP model to support the decision-making approach to risk assessment in real estate development against STEEP factors. Saaty (1999) suggests that Analytic Network Process (ANP) is the systematic approach which is able to deal with both quantitative and qualitative factors under multiple criteria. It is because of this process deal with a multi criterion analysis and comparison, this process outcome are also being in a mathematic statistics format, which could be adopted for further decision making in regard to the risk response and mitigation. A case study of a residential and commercial mixed-used project in Liverpool City centre is used to demonstrate the effectiveness of the ANP model.

2. METHODOLOGY
Methodologies adopted in this research include literature review and face-to-face questionnaires with the real estate practitioners to gain information in regard to current situation in risks assessment for real estate development, following by the data analysis to support ANP model, and case study to test the effectiveness of ANP model to support decision-making in feasibility study for real estate development. A comparison between the existing risk assessment model and ANP is illustrated below:

Figure 1 ANP and Existing Risk Assessment Model This figure illustrates traditional risk assessment method compare with the adoption of ANP method to assess risks in real estate development project. The risks assessment process starts at the selection of an appropriate method, which depend on the characteristics of a specific problem and the preference of decision-makers. Risks classification is necessary to define risk assessment criteria. In the case of using traditional method, the first step is to conduct a panel/board discussion about risks that affect to the project, while each participants use their experience to identify or classify predictable risk events. ioMosaic (2002); Kindinger, 2002 and Rafele, et al., (2005) indicate that the next step is to set up an assessment method, and current practice is mostly to build a matrix for risks assessment. This matrix is therefore used to

describe the likelihood and consequence of each risk in a tabular format. As a result of using the matrix, the panel may find the degree of overall degree of risk events. However, the results derived by matrix assessment method are not based on either non-linear mathematic calculation or objective assumptions related to a real business case. Alternatively, in the case of selection the ANP process, the first step is to develop an ANP model, follows by pair-wise comparison process to form a super-matrix of quantified interdependences between paired criteria and the alternatives of development plan. The results calculated by super-matrix calculation can suit the project team in order to get a numerical suggestion regarding to the most appropriate development plan. The result from ANP is useful to support the decision-making process toward the project risk mitigation. In addition, a project knowledgebase is required to be integrated into the process for using either traditional method or ANP method, in order to complete decision-making tasks. The knowledgebase provides the adequate and accurate information to achieve reliable results, and the knowledge can be collected from existing or new projects.

3. RISK ASSESSMENT CRITERIA

Risks assessment criteria, emphasising on risks and their consequences in real estate development is set up, based on literature review and the researchers experience. These assessment criteria are set up based on Social, Technological, Economic, Environmental and Political factors (STEEP), those affects are necessary when the developers conducting a project feasibility analysis. Since STEEP factor shall cause the variety of risks throughout each project development stage. In this regard, the assessment criteria including each sub-criterion are summarized in the Table 1, this table classifies overall on both quantitative and subjective risks. In addition, it is adopted as the assessment criteria to measure the risks and their impact to the Real Estate Development industry, prior to the Analytic Network Process (ANP) analysis. This table includes five major criterion and their 33 sub-criterion (please see the Table 1 below).

Table 1 Risks Assessment Criteria for the real estate development

Criteria Sub-Criteria Workforce availability Community acceptability Cultural compatibility Public hygiene Site conditions Designers and Constructors Multiple functionality Constructability Duration Amendments Facilities management Accessibility & Evacuation Durability Valuation methods Degree of Developers satisfaction to local workforce market (%) Degree of benefits for local communities (%) Degree of business & lifestyle harmony (%) Degree of impacts to local public health & safety (%) Degree of difficulties in site preparation for each specific plan (%) Degree of Developer satisfaction to their performances (%) Degree of multiple use of the property (%) Degree of technical difficulties in construction (%) Total duration of design and construction per 1,000 days (%) Possibility of amendments in design and construction (%) Degree of complexities in facilities management (%) Degree of easy access and quick emergency evacuation in use (%) Probability of refurbishment requirements during buildings lifecycle (%) Overall value of the Environmental Impacts Index Degree of impacts to use and value due to regional climatic variation (%) Degree of impacts due to increment of loan rate (%) Representative references Danter (2007) Danter (2007) Danter, 2007 NHS Standards Danter (2007) Khalafallah, et al. (2002) Danter (2007) Lam, et al., 2006 Khalafallah, et al. (2002) Khalafallah, et al. (2002) Moss, et al. (2007) Moss, et al. (2007) Chen (2007) Chen, et al. (2005) UNEP (2007) Sagalyn (1990); FSA (2005); Nabarrol & Keys, (2005); FSB (2007); Frodsham (2007) Adair & Hutchison (2005); Frodsham (2007) Adair & Hutchison (2005) Morledge, et al. (2006); FSA (2005); FSB (2007) Adair & Hutchison (2005) http://www.statistics.gov.uk/ D&B (2007); Adair & Hutchison (2005); Gibson & Louragand, (2002) Blundell, et al. (2005); Moore (2006) Lee (2002); Adair & Hutchison (2005) Adair & Hutchison (2005) Frodsham (2007) Booth, et al. (2002) Sagalyn (1990); Watkins, et al. (2004) Arthurson (2001) Gehner, et al (2007) ; FSB (2007) LCC (2008) Crown (2008) Crown (2008)

Social risks

Technological risks

Environmental risks

Adverse environment impacts Climate change

Interest rate

Property type Market liquidity Currency conversion Demand and Supply Purchaseability Brand visibility Capital exposure Lifecycle value Area accessibility Buyers Tenants Investment return Political Groups/Activist Commercial Tax Policy Local Tax Policy Council Approval License Approving

Degree of location concentration (%) Selling rate of same kind of properties in the local market (%) Degree of impacts due to exchange rate fluctuation Degree of regional competitiveness (%) Degree of affordability to the same kind of properties (%) Degree of Developers reputation in specific development (%) Rate of estimated lifecycle cost per 1 billion pound (%) 5-year property depreciation rate (%) Degree of regional infrastructures usability (%) Expected selling rate (%) Expected annual lease rate (%) Expected capitalization rate (%) Degree of protest by the urban communities (%) Rate of Commercial Tax impact (%) Rate of Council Local Tax (%) Total Days of construction, design approval process by Liverpool City Council (LCC) Total Days of license approval process

Economic risks

Political Risks

4. APPLICATION OF ANALYIC NETWORK PROCESS (ANP)

The decision-making model proposed in this paper applies ANP to set up the risk assessment at project feasibility study stage. According to the established risks assessment criteria in Table 1, the ANP model herein based on these 33 defined risk assessment criteria. The model is set up using Super Decisions software for decision-making, created by the Creative Decisions Foundation; and implemented by Professor Thomas Saaty (2005). ANP model comprises 6 clusters and 33 nodes, which are set up accordingly to the criteria and sub-criteria defined in Table 1. The Alternative cluster is used to comprehend alternative plans to be evaluated against risk assessment criteria in a case study; and there are 2 nodes to represent 2 alternative plans for a specific real estate development. ANP method provides an effective mechanism for developers to quantitatively evaluate interrelations between either paired criteria or paired sub-criteria; and this enables the developers to use their expertise to the assessment of all defined risks (see Table 1) occurred in real estate development industry.

Figure 2

ANP Model for real estate development risk assessment

The ANP model, as illustrated in Figure 1, consists of 6 clusters which are Alternatives, Environmental Risks, Social Risks, Economic Risks, and Technological Risks. There are 35 nodes inside this ANP model; amongst them, there are 2 nodes inside the Alternative cluster, which are Plan A, and Plan B represent alternative plans for a specific real estate development in Liverpool, in regard to select the most appropriate plan. The other 33 nodes are located in differenced 5 clusters in accordance with their belongingness to those clusters as described in Table 1. Saaty (2005) describe that two-way and looped arrow lines in Figure 1 describe the interdependences that exist between paired clusters as well as nodes (cited by Chen & Khumpaisal (2008). In addition, there are fixed interrelations between paired clusters, meanwhile there are fixed interrelations between paired nodes inside one cluster as well as from two different clusters. In order to measure all interrelations inside the ANP model quantitatively, the questionnaire survey to compare the relative importance between paired clusters and nodes is required to collect the precise information from the real estate practitioners. According to the questionnaire survey, the experts knowledge and information in each specific domain is collected and concentrated into an ANP model as a result, the ANP model can perform as a decision-making support tool based on knowledge reuse. In this paper, the ANP model is set up by the authors only; and the model will be further developed based on questionnaire survey after a pilot study through the experimental case study. The ANP model illustrated in Figure 1, structuring and quantifying all possible interdependent relations inside the model, pair-wise comparison is adopted using subjective judgements made in regard to fundamental scale of pair-wise judgments. Saaty (2005) (see Table 2). Table 2 generally describes how to conduct pair-wise comparison between paired clusters as well as nodes in regard to their interdependences defined in the ANP model (see Figure 1) and relative importance based on their specific characteristics and experts knowledge. The ANP model is set up based on the risks assessment criteria to make judgments to quantify interdependences for 33 risk assessment criteria inside cluster 2 to 6 (see Figure 1), and specific characteristics of alternative plans, which used to make judgments in quantifying interdependences for alternatives in the experimental case study. Table 2 ANP Judgements between paired clusters/nodes
Clusters/Nodes Cluster I Node Ii
Note: 1. The fundamental scale of pair-wise judgments: 1= Not important, 2= not to moderately important, 3= Moderately important, 4= Moderately to strongly important, 5= Strongly important, 6= Strongly to very strongly important, 7= Very strongly important, 8= Very strongly to extremely important, 9= Extremely important. The symbol denotes item under selection for pair-wise judgment, and the symbol denotes selected pair-wise judgment. I and J denote the number of Clusters, whilst i and j denote the total number of Nodes. The symbol denotes importance initiative between compared Nodes or Clusters.

Scale of pair-wise comparisons 1 2 3 4 5 6 7 8 9

Cluster J Node Jj

2. 3. 4.

5. CASE STUDY AND RESULTS

A case study of a residential and commercial mixed-used project in Liverpool city centre is used to demonstrate the effectiveness of the ANP model in regard to select the most appropriate plan for a specific real estate development project. A case study is conducted based on information collected from an ongoing real estate project in Liverpool City Centre. Some scenarios such as alternative plans in regard to the requirements of comparison study using ANP are made as the assumption. The proposed real estate development locates in central Liverpool with the site area of 40 acres, located between main retail areas, city central business district (CBD), residential areas, walk streets, main roads, and the historical Albert Dock. The Developer partnering with the City Council to revitalise this area for long-term investment in accordance with the Northwest regional and Merseyside Countys economic strategies. To complete the experimental case study purposes, three development plans are considered in this research, which are: Plan A, a retail-led mixed-use inner Liverpool City Centre development; Plan B, a commercial building mixed-use adjacent to inner Liverpool City Centre development. The scenario assumed based on the philosophy of local urban regeneration, which aims to attract more population and customers back to Liverpool City Centre, as well as to maximum utilize of the provided transportation and infrastructures. Mynors (2006) This research employs the face-to-face interview with the practitioner who involved in this studied project to achieve the project information and the developer opinions in regard to the consequential degree of risks affected to his project. The results of interviewing are therefore indicated in Table 3 below. (please see the interviewing summary in appendices)

(grosvenor: 2008) Figure 3 The layout plan of the Experimental Case study

Table 3 Results of alternatives development plan

Criteria Sub-Criteria Workforce availability Community acceptability Cultural compatibility Public hygiene Site conditions Designers and Constructors Multiple functionality Constructability Duration Amendments Facilities management Accessibility & Evacuation Durability Adverse environment impacts Climate change Interest rate Property type Market liquidity Currency conversion Demand and Supply Purchaseability Brand visibility Capital exposure Lifecycle value Area accessibility Buyers Tenants Investment return Political Groups/ activist Commercial Tax Policy Local Tax Policy Council Approval License Approving Unit % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % Alternative Development Plans Plan A Plan B 90 90 60 50 70 50 80 50 60 60 50 50 30 40 95 60 50 60 20 80 30 80 80 80 25 80 50 80 60 75 75 70 100 90 70 80 70 95 50 70 60 80 70 40 70 20 75 30 90 70 70 25 60 70 50 60 50 65 90 100 90 60 70 60 90 50

Social Risks

Technological Risks

Environmental Risks

Economic Risks

Political Risks

Notes PLAN A : Retail-led mixed-use inner Liverpool City Centre PLAN B : Commercial building led mixed-use adjacent to inner Liverpool City Centre

As mentioned above, although interdependences among 33 risk assessment criteria can be measured based on experts knowledge, the ANP model should comprehend all specific characteristics of each alternative plan, which are given in Table 4. According to the fundamental scale of pair-wise judgments (see Table 2), all possible interdependences between each alternative plan and each risk assessment criterion, and between paired risk assessment criteria in regard to each alternative plan are valuated; Table 2 also provides the result of all these pair-wise comparisons, which used to form a two-dimensional super-matrix for further calculation. Saaty (2005) states that the calculation of supermatrix aims to form a synthesized super-matrix to allow for resolution of the effects of the interdependences exist between the nodes and the clusters of the ANP model. (as cited by Chen and Khumpaisal, 2008). In order to obtain useful information for development plan selection, the calculation of super-matrix is conducted following three steps, which transform an initial super-matrix or un-weighted one based on pair-wise comparisons to a weighted super-matrix, and then to a synthesized super-matrix. Results from the synthesized super-matrix are given in Table 4, below.

Table 4 Results Synthesized priority weights Ranking

Comparison or Alternatives development plan results Alternative Development Plans Plan A Plan B 0.6326 0.3674 1 2

According to the results shown in Table 4, Alternative plan A is identified as the appropriate plan for the specific development because it has the highest synthesized priority weight among the 2 alternatives. The differentiate between Plan A and B results indicates the likelihood of the developer to select the appropriate development plan based on results calculated by the ANP method. By the results above, it is suggested that the developer shall select Plan A as the project development plan of the studied project.

10

5. CONCLUSION
An application of Analytic Network Process (ANP) in regarding to assess risks in real estate development when the developer conduct feasibility study stages is introduced in this paper. Risk assessment criteria are set up in order to proceed the ANP calculation was based on literature review and the researchers experiences in real estate development. All assessment criteria are summarised under STEEP factors theory, which could be categorised as: Social, Technological, Environmental, Economic and Political factors. Those factors shall be concerned by the real estate developers and necessary when the project feasibility study conducted prior the project commence. To complete this research, An ANP model is therefore established based on the defined risks criteria associated with STEEP factors. As well as assume that one of two alternative development plans would be selected to develop in Liverpool City centre area. There are 33 risks under five clusters, to ensure a comprehensive coverage of possible risks occurred in real estate development. Then, face-to-face interview with the practitioner involves with this case study has been conducted. According to the results, it can be concluded that the Alternative A The retail-led mixed used property shall be the appropriated development plan, according to the calculation of ANP. Moreover, this Plan A also burdened by the consequence of risk higher that the Alternative B The commercial building led mixed use. According to the results of the interviewing, the practitioners opinion, together with the precise data derived from ANP analysis model. ANP could be concluded as an effective tool to support developers in decision-making based on risks assessment. The ANP model therefore can be adopted by real estate developers in the case of business needs to assess risks in a real estate development. However, the further researches are needed for collecting vast information from the real estate developers in the variety of real estate projects to modify and improve the risk assessment model to suit with the developer requirements in order to improve more consistency and reliable to the risk assessment criteria.

11

REFERENCES
Arthurson, K. (2001), Achieving Social Justice in Estate Regeneration: The Impact of Physical Image Construction, Housing Studies, Vol.16, No.6, pp. 807-826. Adair, A., Hutchison, N. (2005), The reporting of risk in real estate appraisal property risk scoring, Journal of Property Investment and Finance, Vol.23, No.3, pp. 254-268. Bienert, S., Brunauer, W. (2007), The mortgage lending value: prospects for development within Europe, Journal of Property Investment & Finance, Vol. 25, No. 6, pp. 542-578. Blundell, G.F., Fairchild, S., Goodchild, R.N. (2005), Managing Portfolio Risk in Real Estate, Journal of Property Research, Vol.22, No. 2&3, pp. 119-136. Booth, P., Matysiak, G., Ormerod, P. (2002), Risk Measurement & management for real estate portfolios. Investment Property Forum, London. Case, K.E., Shiller, R.J., Weiss, A.N. (1995), Mortgage Default Risk and Real Estate Prices: The Use of Index-Based Futures and Options in Real Estate, Journal of Housing Research, Vol. 7, No.2, pp. 243-258. Chen, Z., Li, H., Wong, C.T.C. (2005), Environmental Planning: an analytic network process model for environmentally conscious construction planning, Journal of Construction Engineering and Management, Vol.131, No.1, pp. 92-101. Chen, Z., Clements-Croome, D., Hong, J., Li, H., Xu, C. (2006), A multicriteria lifespan energy efficiency approach to intelligent building assessment, Energy and Buildings, Vol.38, No. 5, pp. 393-409. Chen, Z. (2007), Multicriteria Decision-Making for the Sustainable Built Environment, The University of Reading, Reading, UK. ISBN: 0-7049-1209-0. Chen, Z., Khumpaisal, S. (2008), An Analytic Network Process for Risks Assessment in Sustainable Commercial Real Estate Development, Journal of Property Investment and Finance, Special Issue: Sustainable Commercial Real Estate CISDM Research Department (2006), The Benefits of Real Estate Investment: 2006 Update, Center for International Securities and Derivatives Markets, Isenberg School of Management, University of Massachusetts, Amherst, Massachusetts, USA. <http://cisdm.som.umass.edu/research/pdffiles/benefitsofrealestate.pdf> (30 Dec. 07). Clarke, J., C. and Varma, S., (1999), Strategic Risk Management : the New Competitive Edge, Journal of Long Range Planning,, Vol. 32, No. 4, pp. 414 - 424. Commission for Healthcare Audit and Inspection (2006) Criteria for assessing core standards in 2006/2007, Commission for Healthcare Audit and Inspection, London. Crown Copyright, (2008), Building Regulations Planning Portal, <http:// http://www.planningportal.gov.uk/england/professionals/en/4000000000001.html>

12

Danter (2007), A Sample Lodging Analysis in the City of Grove, Ohio, Danter Company, Columbus, OH 43215, USA.<http://www.danter.com/PRODUCT/samplodg.pdf> (30 Dec. 07). D&B (2007), Risk Management, Dun & Bradstreet Inc., London, UK. <http://www.dnb.co.uk/Risk_Management/Risk_Management.asp> (30 Dec. 07). Fisher, J.D., Geltner, D.M., Webb, R.B. (1994), Value Indices of Real estate: A Comparison of Index Construction Methods, Journal of Real Estate Finance and Economics, Vol. 9, No. 2, pp. 137-164. French, N., French, S. (1997), Decision theory and real estate investment, Journal of Property Valuation and Investment, Vol.15, No.3, pp. 226-232. Frodsham, M. (2007), Risk Management in UK Property Portfolios: A Survey of Current Practice. Investment Property Forum, London. <http://www.ipf.org.uk/resources/pdf/research/research_reports/Risk_Management_Summary.pd f > (30 Dec. 07). FSA (2005), Strengthening capital standards, Financial Services Authority (FSA), London. <http://www.fsa.gov.uk/pubs/cp/cp05_03.pdf > (30 Dec. 07).

FSB (2007), Financial Stability Report, April 2007, Issue No. 21, The Financial Stability Board (FSB), Bank of England, UK. <http://www.bankofengland.co.uk/publications/fsr/index.htm> (30 Dec. 07). Gehner, E., Halman, J.I.M., Jonge H. de (2006). Risk management in the Dutch real estate development sector: a survey, Proceedings of the 6thInternational postgraduate research conference in the built and human environment, University of Salford, UK. pp. 541-552. Gibson, V.A., Louragand, M. (2002), Risk Management and the Corporate Real Estate Portfolio, Working paper, Centre for Real Estate Research, Business School, The University of Reading, UK. <http://www.reading.ac.uk/LM/LM/fulltxt/0202.pdf> (30 Dec. 07). He, Z. (1995), Risk management for overseas construction projects, International Journal of Project Management, Vol. 13, Issue 4, pp. 231-237. Hendershott, H., P. and Hendershott, J. R. (2002), On Measuring Real Estate Risk, Real Estate Finance (Euromoney Institutional Investor PLC), Winter 2002, Vol. 18, Issue 4, pp. 35- 42. Huffman, F.E. (2002), Corporate real estate risk management and assessment, Journal of Corporate Real Estate, Vol.5, No.1, pp. 31-41. ioMosaic (2002), Designing and Effective Risk Matrix: An ioMosaic Corporation Whitepaper, ioMosaic Corporation, Houston, TX 77057, USA. <http://archives1.iomosaic.com/whitepapers/risk-ranking.pdf> (30 Dec. 07).

13

Khalafallah, A., Taha, M., El-Said, M. (2005), Estimating Residential Projects Cost Contingencies Using a Belief Network, Proceedings of Project Management: Vision for Better Future Conference, Cairo, Egypt, November 21-22, 2005. <http://people.cecs.ucf.edu/khalafal/publications/publications_files/EstimatingContingencies. pdf> (30 Dec. 07). Kindinger, J.P. (2002),"The Case for Quantitative Project Risk Analysis", Tenth Annual Conference on Quality in the Space and Defense Industries, March 4-5, 2002, Florida, USA. <http://www.lanl.gov/orgs/d/d5/documents/case.pdf> (30 Dec. 07). Lee, S.L. (2002), When Does Direct Real Estate Improve Portfolio Performance?, Working paper, Centre for Real Estate Research, Business School, The University of Reading, UK. <http://www.reading.ac.uk/LM/LM/fulltxt/1703.pdf > (30 Dec. 07). Liverpool City Council (2008), Liverpool City Council Property Tax, Liverpool City Council Website, <http://www.liverpool.gov.uk/Council_government_and_democracy/ Council_tax/index.asp> (19 Mar. 08) Morledge, R., Smith, A., Kashiwagi, D.T. (2006), Building Procurement, Blackwell Publishing. Moss, Q.Z., Alho, J., Alexander, K. (2007), Performance measurement action research, Journal of Facilities Management, Vol.5, No. 4, 2007, pp. 290-300. Mynors, P. (2006). What do we need to consider to deliver a more sustainable city? Presentation. CATCH Dissemination Event, Liverpool, 28 - 29 June 2005. Nabarro, R., Key, T. (2005) Performance measurement and real estate lending risk, in Real estate indicators and financial stability, BIS Papers No 21, Bank for International Settlements (BIS), April 2005, pp. 70-90. <http://www.bis.org/publ/bppdf/bispap21.htm> (30 Dec. 07). Nakagawa, T. and Sekitani, K (2004), A Use of Analytic Network Process for Supply Chain Management, Asia Pacific Management Review, Vol. 9, Issue 5, pp. 783-800. Rafele, C., Hillson. D., Grimalai, S. (2005), "Understanding Project Risk Exposure Using the Two-Dimensional Risk Breakdown Matrix", Proceeding papers of 2005 Project Management Institution Global Congress, Edinburgh, Scotland. <http://www.risk-doctor.com/pdf-files/pmi-e-rbmpaper.pdf> (30 Dec. 07). Saaty, T.L. (1999), Fundamentals Of The Analytic Network Process", Proceeding papers of 1999 ISAHP Conference, August 12-14, 1999, Kobe, Japan. Saaty, T.L. (2005), Theory and applications of the analytic network process, RWS Publications, Pittsburgh, USA. Sagalyn, L.B. (1990), Real Estate Risks and the Business Cycle: Evidence from Security Markets, The Journal of Real Estate Research, Vol. 5, No. 2, pp. 203-220.

14

Strischek D. (2007) Regulatory guidance on real estate risk: mind the gap for great guidance on good lending, The RMA Journal, April 2007. Titarenko, B.P. (1997), Robust technology in risk management, International Journal of Project Management, Vol. 15, Issue 1, pp. 11-14. UNEP (2007) Buildings and Climate Change - Status, Challenges and Opportunities, United Nations Environment Programme (UNEP), Nairobi, Kenya. Watkins, C.J., Hughes, S.C., Sims, R., Hildebran, M.E., Hoyer, B.D. (2004), Assessing Real estate Portfolio Risk, Supervisory Insights, Federal Deposit Insurance Corporation, Washington, USA. <http://www.fdic.gov/regulations/examinations/supervisory/insights/cre_lending.html> (30 Dec. 07).

Wheation, C., W., Torto, G., R., Sivitanides, S., P., Southard, A., J., Hopkins, E., R. and Costello, M., J. (2001), Real Estate Risk: A Forward-Looking Approach, Real Estate Finance (Euromoney Institutional Investor PLC), Fall 2001, Vol. 18, Issue 3, pp. 20- 30. Younes, E., Kett, R. (2007), Hotel investment risk: what are the chances? Journal of Retail Leisure Property, Vol.6, No.1, pp. 69-78.

15

APPENDICES

1. Summary of Interview

2. The example of face-to-face interview questionnaires

16

SUMMARY OF INTERVIEW
Date and Time : 30th April, 2008, interviewing started at 10.05 am. Finished at 10.40 am. , Estimated duration 35 minutes Meeting room of the Developer, 2nd Floor, Cunard Building, Water street, Liverpool, L3 1JR Mr. Martin J. Jennings, Partner of the well-known, global wide project management consultancy services firm. (herein after named the developer) Sukulpat Khumpaisal, Research Student, School of Built Environment, Faculty of Technology and Environment, Liverpool John Moores University Risks Assessment in Real Estate Development: an Analytic Network Process approach.

Venue

Interviewee

Interviewer

Topic of Interview:

Objectives of this interview : 1. To investigate the Developers opinion in regard to the consequences of risks impact to the real estate development plans. 2. To degree the level of each risks impact to the assumed development plan. 3. To complete the research work of the interviewer, research title Risks Assessment in Real Estate Development: an Analytic Network Process approach. Results gain in this interview would be analysed in Analytic Network Process (ANP) model. 4. To pilot study of established risk assessment criteria, in the interviewers further thesis. Interviewing structures : An informal interviewing in casual atmosphere, this interviewing is facilitated by the face-to-face questionnaires (see attaches). Questionnaires contains 5 risk major criteria and 33 sub-criteria, evaluating method of each sub-criteria including the assumed alternative plan. Note: In order to protect the companys reputations and the interviewees privacy and his own confidential. This research will not specify the interviewees name hereinafter in this paper. As well as no media-recording during the interview period.

17

Interview Summary
The interview starts with the introduction of the research topic, and the objectives of interviewing. To accomplish the objectives of this research, 2 development plans has been assumed as the Developer involved project, which are Plan A: Retail-led mixeduse inner Liverpool City Centre and Plan B: Commercial building led mixed-use adjacent to inner Liverpool City Centre. Lists of criteria are defined by the researcher, and being asked to understand the interviewees opinions as he is involved as real estate developers role. The summation of interview is thus finding and showing in the followings paragraphs: 1. Social Risks This criteria mode contains with 4 sub criteria, which are Workforce availability, Community acceptability, Cultural compatibility and Public Hygiene. He stated that the workforce availability is the main issue and has a big impact for the developers in term of lack of good-skilled construction labours or qualified contractors. He weights 90% for both alternative plans in regard to the consequences of risks from this sub-criterion. According to his opinion, he insists that Retail-led properties are impacted by the social risks than the commercial properties in regard to the community acceptability and cultural compatibility. Especially on the Public Hygiene factor, it is found the significant differentiate between both development plans, as 80% in Retail Mixed-use comparing to 50% in Commercial led. It is according to the amount of Retail Mixeduse s debris such as rubbish from restaurants or food stalls, or load of waste water, sewerage are much more than the Office buildings, and those would impact to the overall public hygiene of the Liverpool city communities. 2. Technological risks Nine sub-criteria are containing within this risks mode, the Technological risks are the risks almost occurred in the project construction and property management stages, those could be summarised as: Site conditions; Designers and Constructors; Multiple functionality; Constructability; Duration; Amendments; Facilities management; Accessibility & Evacuation and Durability. Based on the developer opinion, the degrees of risks affect to the development plan are quite similar in each sub-criterion, except the Multiple- functionality factor. For the Alternative A, the percentage of risks is 30%, meanwhile 70% in Alternative B, he states that it because of the commercial real estate building is designed or planned for a specific leasing purpose only, it structure or service systems is difficult to modify to suit with the other purposes. Based on his opinion, he insists that Duration is the most important factor in this Technological mode, as seen in the results, he states 95% for alternative A and 80% for B, since it affect to the developers, in term of investment return and the payment schedule, the longer development duration, equal to the higher cost of workforce and more suffer from the interest payment to the loan lenders.

18

3. Environmental risks There are only 2 sub-criteria contained within this mode, the researcher aims to find the developer opinion in regard to impact of Adverse environment impacts and Climate change to both alternative development plans. Interview results show that the developer has a highly concern in risk caused by the impacts of adverse environment, these also reveal that retail-led properties are suffered by this adverse environment than the commercial-led properties, as 95% comparing with 80%. Whilst the results of Climate change show the similar 20% for both alternatives. 4. Economic risks Economic risk is the most concerned risk factor, while any developer develops their real estate development project. It is due to the income generated characteristic of real estate development property. According to this concern, there are 13 sub-criteria contained within the Economic risks mode, the selected sub-criteria are categorised based on the risks caused by financial issues such as Interest rate, Currency conversion, Expected selling and annual lease rate, and Investment return. This risks mode also covering on the risks caused by marketing factors, which are Property type (location concentration), Market liquidity, Demand and supply (the competitiveness of properties), Purchase ability and Brand Visibility. Finally, the researcher also concern on risks caused by the property value and its depreciation, the sub-criteria to be concerned are such; Capital exposure (life cycle cost per 1 billion pound), Life cycle value and the Area accessibility. The developer states that real estate developers generally concern the risks caused by the financial issues since those directly affect to the projects vitality in term of income and project cash-flow. As seen in the survey results, he indicates higher percentage for both development alternatives, mostly higher than 70% in each indicator. However, the impact of Currency conversion are lower, since the real estate developers mostly concern in this issue and prepare the contingency plan to mitigate the consequence of the currency fluctuation. Risks caused by marketing issues also being concentrated by the developer, he indicates the higher percentage in this group as well. Especially for the alternative plan A, retailled property are strongly affected by risks caused by location concentration, market liquidity, demand and supply as well as risks of same properties types competitiveness. The developer also notes that the retail-led properties in Liverpool city centre might be strongly affected by the competitive from retail shopping centres in Manchester core business area, because of the short transportation distance and the products are variety than the retail shopping in Liverpool. Finally, in regard to the property life cycle value and property depreciation, the developer gives the attracted opinion that retail-led properties are affected by the physical deteriorated than the commercial properties in term of customers and functional usage including the standards of properties maintenance.

19

5.

Political Risks The last risk assessment group is concerning in risks caused by the political issues, in this research, political risks have been limited in the local or the community level only, not covering on the risks in national government level. For example, this research only scoped on the risks caused by the protested of the local political/activist groups, the tax affect both commercial and local tax, and the Liverpool city councils planning approval procedures for the development. He indicates that political risks are the high issues affected to both development alternative plans. Especially, the risk caused by the delay in approval procedure are highly impact to overall project management stage in term of project cost and return of investment, he degrees 95% for alternative A and 90% for alternative B. The developer also suggests that in order to mitigate the consequence of risk, the developer shall prepare the contingency plan to response to this kind or risk or attend more concentrations in each city council planning procedures.

Conclusions and suggestions After finishing the interview in regards to the established risks assessment criteria, the developer has suggest the additional criteria in order to complete overall risk assessment criterion. Focus on the Technological risks, he states that Land availability shall be added into the criteria, it is because of the scarcity of land in UK context, and the variable of the land prices, both are affected directly to the condition of development site and project income. This suggestion is useful for the further research to modify the assessment criteria to be suited with the practitioners requirements. In addition, based on his opinion, he indicates that the highest risk mode that shall be attended by real estate developers are the risks in Economic mode, due to the current economic situation (1st quarter, 2008), there are many signs showing the impact of Economic risks, for example, fuel prices increase, the land and property price fall. However, if we trace back to the started time of Liverpool centre regeneration project (2006), he considers that Political risks had the higher impacts to this project, in term of council approval procedures. Finally, he also suggests some good case studies in Liverpool area, for example the new Everton football club stadium, because this project is currently suffered from variety of risks, particularly risks from planning and protested groups.

20

Interview details (categorised in each risk mode) Social Risks

Sub-criteria Valuation Methods Unit % % % % % Alternative Plans A B 90 50 50 50 90 60 70 80

Workforce availability Community acceptability Cultural compatibility Public hygiene

Degree of Developers satisfaction to local workforce market Degree of benefits for local communities Degree of business and lifestyle harmony Degree of impacts to local public health & safety

Technological Risks
Sub-criteria Valuation Methods Unit % Alternative Plans A B

Site conditions Designers and Constructors Multiple functionality Constructability Duration Amendments Facilities management Accessibility & Evacuation
Durability

Degree of difficulties in site preparation for each specific plan Degree of Developer satisfaction to their performances Degree of multiple use of the property Degree of technical difficulties in construction Total duration of design and construction per 1,000 days Possibility of amendments in design and construction Degree of complexities in facilities management Degree of easy access and quick emergency evacuation in use
Probability of refurbishment requirements during buildings lifecycle

% % % % % % % %
%

60 50 30 40 95 60 50 60
20

60 50 70 60 80 70 40 70
20

21

Environmental Risks
Sub-criteria Valuation Methods Unit % Alternative Plans A B

Adverse environment impacts Climate change

Overall value of the Environmental Impacts Index Degree of impacts to use and value due to regional climatic variation

% %

80 30

75 30

Economic Risks
Sub-criteria Valuation Methods Unit % Alternative Plans A B

Interest rate Property type Market liquidity Currency conversion Demand and Supply Purchaseability Brand visibility Capital exposure Lifecycle value Area accessibility Buyers Tenants Investment return

Degree of impacts due to increment of loan rate Degree of location concentration Selling rate of same kind of properties in the local market Degree of impacts due to exchange rate fluctuation Degree of regional competitiveness Degree of affordability to the same kind of properties Degree of Developers reputation in specific development Rate of estimated lifecycle cost per 1 billion pound 5-year property depreciation rate Degree of regional infrastructures usability Expected selling rate Expected annual lease rate Expected capitalization rate

% % % % % % % % % % % % %

80 80 80 25 80 50 80 60 75 80 70 100 90

90 70 70 25 60 70 50 60 50 60 90 100 90

22

Political Risks
Sub-criteria Valuation Methods Unit % Alternative Plans A B

Political Groups/Activist Commercial Tax Policy Local Tax Policy Council Approval License Approval

Degree of protest by the urban communities Degree of Commercial Tax affect to project Rate of Council Local Tax Total Days of construction, design approval process by the Liverpool City Council (LCC) Total Days of license approval process

% % % % %

70 80 70 95 50

60 70 60 90 50

23

24

School of Built Environment LIVERPOOL Sukulpat Khumpaisal (Pat) E-mail: S.Khumpaisal@2007.ljmu.ac.uk ; Phone: 07942-396823 Questionnaire Survey Risks Assessment in Real Estate Development: an Analytic Network Process approach

Please estimate risks that may be relevant to alternative assumptions (Plan A and B) or your real project plans. Thanks.
Alternative Plans* A B
90 60 70 80 60 50 30 40 95 60 50 60 20 80 30 80 80 80 25 80 50 80 60 75 80 70 100 90 70 80 70 95 50 90 50 50 50 60 50 70 60 80 70 40 70 20 75 30 90 70 70 25 60 70 50 60 50 60 90 100 90 60 70 60 90 50

Criteria

Sub-Criteria Workforce availability

Valuation methods Degree of Developers satisfaction to local workforce market Degree of benefits for local communities Degree of business and lifestyle harmony Degree of impacts to local public health & safety Degree of difficulties in site preparation for each specific plan Degree of Developer satisfaction to their performances Degree of multiple use of the property Degree of technical difficulties in construction Total duration of design and construction per 1,000 days Possibility of amendments in design and construction Degree of complexities in facilities management Degree of easy access and quick emergency evacuation in use Probability of refurbishment requirements during buildings lifecycle Overall value of the Environmental Impacts Index Degree of impacts to use and value due to regional climatic variation Degree of impacts due to increment of loan rate Degree of location concentration Selling rate of same kind of properties in the local market Degree of impacts due to exchange rate fluctuation Degree of regional competitiveness Degree of affordability to the same kind of properties Degree of Developers reputation in specific development Rate of estimated lifecycle cost per 1 billion pound 5-year property depreciation rate Degree of regional infrastructures usability Expected selling rate Expected annual lease rate Expected capitalization rate Degree of protest by the urban communities Degree of Commercial Tax affect to project Rate of Council Local Tax Total Days of construction, design approval process by the Liverpool City Council (LCC) Total Days of license approval process

Unit % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %

Social Risks

Community acceptability Cultural compatibility Public hygiene Site conditions Designers and Constructors Multiple functionality Constructability

Technological Risks

Duration Amendments Facilities management Accessibility & Evacuation Durability

Environmental Risks

Adverse environment impacts Climate change Interest rate Property type Market liquidity Currency conversion Demand and Supply Purchaseability

Economic Risks

Brand visibility Capital exposure Lifecycle value Area accessibility Buyers Tenants Investment return Political Groups/Activist Commercial Tax Policy

Political Risks

Local Tax Policy Council Approval License Approval

Alternative development plans: Plan A: Retail-led mixed-use inner Liverpool City Centre; Plan B: Commercial building led mixed-use adjacent to inner Liverpool City Centre.

School of Built Environment LIVERPOOL Sukulpat Khumpaisal (Pat) E-mail: S.Khumpaisal@2007.ljmu.ac.uk ; Phone: 07942-396823 Questionnaire Survey Risks Assessment in Real Estate Development: an Analytic Network Process approach