Institute for Transport Studies FACULTY OF ENVIRONMENT Appraisal

Institute for Transport Studies FACULTY OF ENVIRONMENT Appraisal Practice; The Crossrail Approach Tom Worsley

Crossrail – the scheme Crossrail map with connections Journey time savings Liverpool St - Heathrow 23 mins(55 now), Paddington - Canary Wharf 14 mins (30 now) Addition to London’s rail capacity – overall 6%, C. Area 10% -

Crossrail – from first proposal to start of works 1944 Abercrombie Plan 1987 -9 CLRS • E-W and N-S Crossrail – best options • Docklands - a different type of scheme – referred to ELRS – JLE 1999 1990 Decision deferred – recession and decline in demand 2002 CLRL established by SRA and Tf. L 2005 Montagu review of 2003 Business Case 2005 Updated business case – WEBs 2007 -10 Supplementary Business Rate – 2007 -10 2005 -8 Crossrail Bill/royal assent 2009 Start of works

Standard Appraisal and modelling practice - CLRS 1987 -9 Models and appraisal – (required to follow Df. T guidance) • LTS and RAILPLAN – Fixed employment, fixed trip totals - route, mode and for non-work trips destination choice allowed to change. Delta GC for appraisal. • Conventional benefits plus weights on time savings for interchange, walk, wait and crowding. Road decongestion externalities. • BCRs of between 1. 6 and 1. 9 for the better options – E-W and N-S Crossrail Unresolved • Funding • Institutional arrangements/governance • Scheme champion • Labour supply effects – zonal job totals fixed Exogenous – recession and decline in demand – Crossrail proposal deferred.

Links to Docklands 1987 -92 ELRS – Jubilee Line Extension • Modelling and appraisal as for CLRS – extreme crowding in do-minimum • BCR<1. 0 but scheme approved – completed in 1999 • In place • Funding (subsequently unravelled) • Institutional arrangements to deliver scheme and land use changes (London Transport, LDDC) and measures to induce development • Scheme champion (PM’s support, private developers and LDDC) • LDDC’s forecasts shifted jobs to Docklands zones, but no overall effect on labour supply • Modelling and appraisal methods failed to capture regeneration effects • Very different circumstances from C. London proposals

Transport and the Economy 1999 -2005 Increasing policy focus on transport and the economy - Df. T and HMT interest • SACTRA 1999 – identified links, focus on imperfect markets in transport and land use • Eddington 2005 – Noted good BCRs for transport, changes to Df. T delivery processes recommended Well established appraisal and modelling methods – incremental changes Dependence on data, evidence and models to capture wider benefits Main developments • Economic Impact Report – regeneration in policy priority areas • Wider Economic Benefits

Wider Economic Impacts in Df. T Appraisal Guidance -2005 Wider Economic Impacts – extension of standard CBA • Agglomeration • Increased competition between firms (not applicable in England) • Imperfect competition • Labour supply • Increased participation • Move to more productive jobs Now used for most larger schemes – requirement from 2012

Agglomeration Benefits • A measure of the proximity of businesses and of workers to each other resulting in external benefits • Based on evidence of effective density and productivity by sector • Delta ED as estimated from transport model’s delta GC for commuting and business and zonal employment data/forecasts • Data /forecasts of value added per worker in Df. T guidance • Coefficients defined in guidance • Distance decay parameter • Productivity wrt ED • Requires multi-modal model compliant with Df. T guidance

Labour Supply Effects Increased participation • Labour supply elasticity wrt post tax wages plus change in generalised cost of journey to work • Welfare benefit – additional tax receipts • GDP effect – gross output from increase in numbers working Move to more productive jobs (sensitivity test) • LUTI model to estimate changes in location of employment • Data on output per worker by zone • Welfare and GDP effects as above

Appraisal and decision rules in Df. T No absolutes, but the Department has; • Published a definition of value for money • Provided categories based on adjusted BCRs which define poor, low medium, high and very high vfm • Made use of these categories for ranking and sifting out all poor, most low and many medium schemes • Adopted HM Treasury’s 5 case business model, setting the economic case into context

Crossrail and Wider Economic Benefits Proposal reviewed in 2003 following upturn in commuting and approved in 2008 • All the now standard transport user benefits plus (post 2005) WEBs adding around 50%+ to user benefits • No LUTI model – delta capacity constraint redistributed workers to C. London • What had changed since CLRS? • Institutions/governance – Montague • Funding – WEBs helped • High vfm – new Df. T guidance includes WEBs • Champion – changed institutions

Composition of Crossrail benefits Category of benefit £bn 2002 prices Df. T Public transport users; commuting and leisure 6. 1 Public transport users - business 4. 1 Road users; commuting and leisure 1. 6 Road users; business 0. 6 Indirect tax change -1. 4 Total transport user benefits 11. 0 Agglomeration 3. 1 Move to more productive jobs 2. 0 Labour force participation 0. 8 Imperfect competition 0. 5 Total Wider Impacts 6. 4 Tf. L 15. 5 7. 0 -18. 0

Alternative metrics/objectives GDP per £’s worth of public spending • GDP defined as PV of generalised cost savings for ‘productive’ trips, defined for this purpose as commuting, freight and business • Used by Df. T ministers but • Not a pass/fail test – at the best sets priorities • Identification of GDP/non-GDP very crude • Impossible to audit/evaluate • Not in line with National Accounts

An alternative approach – GVA effects KPMG approach makes estimates of • Productivity effects of increased agglomeration • Changes in the distribution of employment in response to increased accessibility • Wage equation – elasticities of 0. 11 business to business and 0. 09 labour to business for wage rate wrt generalised cost • 80% of this productivity gain explained by changes in sectoral mix; 20% changes in productivity within sectors • Redistribution through an elasticity of employment density wrt labour and business accessibility – some allowance for intraregional shift.

Conclusions and Challenges Is CBA plus WEBs an adequate guide for decision-makers? Issues • Translating time savings into real effects • Is the BCR/Vf. M the right metric? • Appraisal of land use and other changes to deliver the overall outcome • Identifying beneficiaries – potential funding sources • Optimism bias in forecasts • Does it come early enough to influence decisions? In support of CBA plus WEBs • Has worked for most schemes – exceptions – JLE, HS 2 • Provides consistency between schemes and between forecasts and of appraisal methods • It relies on the transport model and so has low costs