ET2050 European Territorial Scenarios modelled by SASI Klaus Spiekermann and Michael Wegener ET2050 Project Group Meeting Barcelona, 25-27 September 2013.

ET2050

European Territorial Scenariosmodelled by SASI

Klaus Spiekermann and Michael Wegener

ET2050 Project Group MeetingBarcelona, 25-27 September 2013

2

The SASI Model

3

The SASI model

There are three methods to model the impacts ofpolicies on regional economic development:• Multiplier effects of infrastructure invest

ments (Aschauer, 1993)• Regional production functions incorporating

infrastructure as production factor (Jochimsen, 1966; Biehl, 1986, 1991)

• Interregional trade flows as a function of interregional transport costs (Peschel, 1981; Bröcker, 1995) and input output linkages (Echenique, 1990) and economies of scale (Krugman, Venables,1995)

SASI

4

The SASI model

The SASI model is a recursive -dynamic simulationmodel of socio economic development of regions inEurope under assumptions about • European economic development and

external net migration,• European transport policies (TEN-T), • regional subsidies (ERDF, EAFRD, ESF, CF).

The SASI model differs from other regional economicmodels by modelling not only production (thedemand side of regional labour markets) but alsopopulation (the supply side of labour markets) and travel and freight transport flows.

5

Regional production function

In state- of- the- art models of regional development based on production functions the classic production factors land, labour and capital are replaced by location factors, such as:• Economic structure• Productivity• Accessibility• Labour supply• Services• Settlement structure• Research and development• Education• Quality of life

6

Regional production function

Extended production function:

where Ai is potential accessibility:

...

iiiii AKRLQ Production

AccessibilityLand

Labour Capital

Others

Travel cost betweenregions i and j

j

ijji cWA )(exp

Destinationsin region j

Accessibilityof region i

7

Regional net migration function

Net migration function:

where

• qi(t–3) is GDP per capita of region i• q(t-3) is average European GDP per capita• vi(t–3) is quality of life of region i• v(t-3) is average European quality of life … all lagged by three years

5.1)3(

)3(5.1

)3(

)3()(

tv

tv

tq

tqtm ii

i

Attractiveness as place to work

Attractivenessas place to live

Net migrationof region iin year t

8

GDP

AccessibilityProduction

function

Employment

Migrationfunction

PopulationIncome

Labourforce

Unemployment

SASImodel Transfer

policiesTransportpolicies

9

SASImodel

RegionsRegions

10

The SASI Model in ET2050

11

Model developments for ET2050

• Conversion of the region system to the 2006 NUTS 3 classification

• Extension of forecasting horizon from 2030 to 2050

• Extension of study area to the ESPON Space (EU27+4)

• Development of a simple model of regional and long- distance travel and freight transport

• Calculation of environmental indicators as energy consumption and CO2 emissions of transport

12

Model integration in ET2050

• Adjustment of exogenous assumptions to those of MASST and MULTIPOLES.

• Translation of the NUTS-2 typology of regions pro-posed in the MASST to NUTS-3 regions.

• Adoption of the exploratory scenarios A, B and C defined for MASST and MULTIPOLES.

• Provision of NUTS-3 model results for the three exploratory scenarios to METRONAMICA.

13

Questions answered

How will different spatial orientations of European

• infrastructure investments (TEN -T)• regional subsidies (ERDF, EAFRD, ESF, CF)

affect

• regional economic development, • regional population/migration,• interregional travel and freight flows,• territorial cohesion and polycentricity, • energy consumption/CO2 emissions?

14

Baseline Scenario

15

Baseline Scenario

The Baseline Scenario for 2030 and 2050 is basedon BAU assumptions about • European economic development,• European net migration,• European regional subsidies,• European transport policies.

and produces forecasts of:• regional economic development, • regional population/migration,• interregional travel and goods flows,• energy consumption/CO2 emissions, • territorial cohesion and polycentricity.

16

Baseline Scenario

Populationmaximumin 2035:BaselineScenarioassumption:population1981-2051

17

Baseline Scenario

Immigrationconstrained:BaselineScenarioassumption:EU27 netmigration per year(1,000)1981-2051

18

Baseline Scenario

"Sluggishrecovery":BaselineScenarioassumption:GDP (Euroof 2010)1981-2051

19

Baseline Scenario

Subsidiesgrow withEU GDP:BaselineScenarioassumption:StructuralFunds (billion Euroof 2010)1981-2051

20

Baseline Scenario

Subsidiesmainly to poorerregions:Baseline Scenario assumption:StructuralFunds as% of GDPv. GDP per capita1981-2051

21

Baseline Scenario

Risingenergy prices:BaselineScenarioassumption:Oil priceper barrel(Euro of 2010)1981-2051

22

Baseline Scenario

Growingtransportenergyefficiency:BaselineScenarioassumption:MJ per pkm/tkm1981-2051

23

Baseline Scenario

Cleanervehiclesand morerenewable energy:Baselinescenarioassumption:CO2 emissionof transportper MJ1981-2051

24

Exploratory Scenarios

25


The definition of the SASI exploratory scenarios is based on the same region typology as used by the MASST and MULTIPOLES models but translated into NUTS-3 regions:•In the MEGAs Scenario A large European metro-politan areas are promoted in the interest of com-petitiveness and economic growth.•In the Cities Scenario B major European cities are promoted in order to strengthen the balanced poly-centric spatial structure of the European territory.•In the Regions Scenario C rural and peripheral regions are promoted to advance territorial cohesion between affluent and economically lagging regions.

26


In the SASI exploratory scenarios A, B and C the assumptions about total European development and European net migration remain the same as in the Baseline Scenario.

However, the exploratory scenarios differ in their assumptions about •the allocation of EU Structural Funds subsidies (see next slide),•European transport policies (see three following slides).

27

Baseline Scenario

The

A (MEGAs)

B (Cities)

C (Regions)

1.0 %

0.5

0.25%

of total EUStructuralFunds

Exploratoryscenarios:Structural Funds

28

Scenario A:Networkimprovements(if necessary)

MEGA

Connections between MEGAs not more than 500 km apart.

Minimum speed:

Road: 90 km/hRail: 200 km/h

29

Baseline Scenario

City

Connections between cities not more than 300 km apart.

Minimum speed:


Scenario B:Networkimprovements(if necessary)

30

Baseline Scenario

Region

Connections between regions not more than 200 km apart.

Minimum speed:


Scenario C:Networkimprovements(if necessary)

31

Exploratory scenario results

32

Baseline Scenario:Accessibility travel road/rail1981















33

GDÜ per capita(1000 € of 2010)

BaselineScenario:GDP percapita2051

34

Difference toBaseline Scenario(%) 2051

Scenario A:GDP per capitaDifferenceto BaselineScenario2051

35


Scenario B:GDP per capitaDifferenceto BaselineScenario2051

36


Scenario C:GDP per capitaDifferenceto BaselineScenario2051

37

Results (2)

Populationdensity(pop/sqkm)

BaselineScenario:Populationdensity2051

38


Scenario A:PopulationDifferenceto BaselineScenario2051

39


Scenario B:PopulationDifferenceto BaselineScenario2051

40


Scenario C:PopulationDifferenceto BaselineScenario2051

41

CO2 emission(t/capita/year)

BaselineScenario:CO2 emission2051

42

Scenario variants

43

Scenario variants

In addition, the Baseline Scenario and the exploratory scenarios A, B and C are combined with alternative framework conditions:

1.Economic decline. Globalisation and growth of emerging economies will lead to stagnation and almost decline of the European economy

2.Technology advance. New innovations in production and transport technology will result in significant growth in labour and transport productivity.

3.Energy/climate. Rising energy costs and/or greenhouse gas emission taxes will lead to strong increases of production and transport costs.

44

Scenario variants

The combination of the exploratoy scenarios and the variants leads to nine additional scenarios:

45

Scenario variants

Economic decline

It is assumed that in Scenarios A1, B1 and C1 total GDP of EU27+4 will grow by only 0.62 per cent p.a. on average between 2011 and 2051 compared to 1.50 per cent in the Baseline.

As in the Baseline Scenario, it is assumed that growth rates will gradually decrease after 2030.

46

Scenario variants

Technology advance

It is assumed that in Scenarios A2, B2 and C2 labour productivity, i.e. GDP per worker, will grow by 1.94 per cent p.a. on average between 2013 and 2051 compared to 0.94 per cent in the Baseline Scenario.

It is assumed that productivity will gradually con-verge between countries towards 2051.

It is assumed that energy efficiency of transport will increase by 0.75 % per cent p.a. compared to 0.45 % per cent in the other scenarios.

47

Scenario variants

Energy/climate

It assumed that in Scenarios A3, B3 and C3 fuel costs of road vehicles will increase by 5 % per cent on average between 2013 and 2051 compared to 1.5 per cent in the Baseline Scenario.

This will result in an average fuel price of 10.20 Euro of 2010 in 2051 compared with 3.00 Euro in the Baseline Scenario. Energy cost of rail transport is assumed to increase by 2 per cent p.a. between 2013 and 2051.

48

Scenario variants

Scenario variant assumtions: summary

1 withour generative effects

49

Scenario variants results

50

Scenario comparison

Transportnetworks &travel cost:Accessibilityof travel byroad/rail1981-2051

51

Scenario comparison

RemainingEast-Westgap: GDPper capita EU15/EU12(1,000 Euroof 2010)1981-2051

52

Scenario comparison

Decliningoverallregionaldisparities: Ginicoefficientof GDP percapita1981-2051

53

Scenario comparison

Persistenturbanstructures in EU15Nationalpolycentricityindex1981-2051

54

Scenario comparison

Dynamic urban structures in EU12:Nationalpolycentricityindex1981-2051

55

Scenario comparison

Increasein energy efficiencyv. growth in volume:Energyconsumptionby transportper capitap.a. (MJ)1981-2051

56

Scenario comparison

Increase in energyefficiency& share ofrenewableenergy:CO2 emissionby transportper capita p.a. (t)1981-2051

57

Conclusions

58

Summary comparison

59

Conclusions

The comparison of scenarios with respect to the three major EU goals gives a straightforward result:•Competitiveness: The A scenarios (MEGAs) produce the largest growth in GDP. The C scenarios (Regions) perform worst in terms of overall economic growth. The B scenarios (Cities) lie in between.•Cohesion: The C scenarios perform best in terms of cohesion and polycentricity. The A scenarios slow the convergence down. The B scenarios lie in between.•Sustainability: The B scenarios are most successful environmentally. The A and C scenarios use more energy and emit more CO2 for transport.

60

Conclusions

The results of the scenario simulations with the SASI model can be summarised as follows:•Promotion of metropolitan areas will maximise economic growth but increase spatial disparities and environmental damage.•Promotion of rural and peripheral regions will increase spatial cohesion but reduce economic growth and sustainability.•Promotion of large and medium-sized cities is a rational trade-off between competitiveness and cohesion and will be best for the environment.

61

Conclusions

These results validate the balanced polycentric spatial organisation of Europe as suggested by the European Spatial Development Perspective (ESDP) and the Territorial Agenda (TA).

The B scenarios (Cities) should therefore be taken as the point of departure for the territorial vision.

62

Next steps

A modelling exercise is permanent work in progress.Therefor the SASI team is looking forward to receiving critique and suggestions for improvement.

There are several possibilities for complementing and improving the SASI model:•combining assumptions of scenario variants?•more comprehensive energy/climate variant?•also publish results for the Western Balkan?•any other?

63

Meta Analysis

in a final step the models used in ET2050 will be com-pared in a meta analysis.

A meta analysis is a way to cross-validate the results of different models by systematically comparing their results to identify differences between them and, if they differ, explore the reasons why.

To overcome the differences in spatial resolution and time horizon between the models, not absolute values but differences between the exploratory scenarios and the Baseline Scenario will compared.

64

Wegener, M., Bökemann, D. (1998): SASI Model: Model Structure. Berichte aus dem Institut für Raumplanung 40. Dortmund: Institute of Spatial Planning, University of Dortmund. http://www.raumplanung.uni dortmund.de/irpud/ fileadmin/irpud/content/documents/publications/ber40.pdf.

Wegener, M. (2008): SASI Model Description. Working Paper 08/01. Dortmund: Spiekermann & Wegener Stadt und Regionalforschung. http://www.spiekermann- wegener. de/mod/pdf/AP_0801.pdf.

Spiekermann, K. Wegener, M. (2013): The SASI Sce-narios until 2050. Project Report for the ESPON-Projekt ET2050 (Territorial Scenarios and Visions for Europe).

More information

ET2050 European Territorial Scenarios modelled by SASI Klaus Spiekermann and Michael Wegener ET2050 Project Group Meeting Barcelona, 25-27 September 2013.

Documents

baseline scenario slide

sasi slide

sasi model slide

baseline scenario assumption

et2050 slide

baseline scenario subsidies

baseline scenario immigration

emissions of transport