Kaldor DT Model

-Open economy -Imperfect Competition in product markets -Dynamic increasing returns to scale due to 1- Intra-industry specialisation (Smith 1778) 2- Inter-industry specialisation (Young 1928) - where whole new industries start to appear. There is an interactive process of AS and AD within the economy.

- y* = responsiveness of output growth to increase in exports - n* = Responsveness of exports to price competition. - h* = responsiveness of labour productivity to growth - the verdoon coefficient - delta - the income levels of other countries - E* - the non-price competitiveness as determined by the elasticity of income.

An Increase in output growth, Dynamic increasing returns by output growth drving productivity Productivity driving competitiveness Competitiveness (international) driving export-demand growth

The model in equations: 1- The export-base relationship: Y₁ = γx₁ 2- Export Demand Growth X₁ = - ηπ_t-1 + δπ_c + εγ_c 3- Regional Price Inflation Π_t= w + τ – r_t 4- Verdoon’s law 1949 R_t = r_e + λy_t

A region can permamently have different growth rates (like exogenous solow model without level effects). A region will grow faster if: - Inflation in competing regions is high -Income growth in export markets is fast -If non-price competitiveness is high -If dynamic increasing returns are strong.

1- a) Encourage diversification into exports that have a competitive advantage b) Create alternative conditions to attract FDI: - Stable political climate - Strong legal system, well protected PRs - Good quality human capital - good infrustructure 2- Restrain nominal wage growth through improved relative price competitiveness: -Tackle union power - assess benefits Empirically relative price competitiveness is less important than non-price competitiveness.

The verdoon law conceals an exceedingly complex process. The ways in which the expansion in output leads to an improvement of productivity growth are poorly understood. Verdoonn (1949) argued that rapid output growth creates opportunnities for greater division and specialisation of labour. Other contributory factors include the tendency for technical progress to be stimulated when output grows rapidly The empirical evidence offered in support of the Verdoon relationship is controversial. - Internal economies of scale only play a minor role in growth-pole expecttations of regional growth. A more convincing explanation of cumulative growth stressed by Krugman 1991 incorporates the effect of external economis of scale into the explanation of regional growth disparities. Thus other explanations for the cumulative growth process which arent as concealing as the Verdoon coefficient are the localisation economies and agglomeration economies.

McCombie and De Ridder 1984 find firm statistical support for the Verdoon law More recently Fingleton and McCombie 1998 have estimated (across 178 regions in the EU) that the Verdoon coefficient lies between 0.5 and 0.6 which indicates that rapid output growth creates opportunities for greater division and specalisation of labour. (for every 1% increase in a region's output, there is a 0.5% increase in the growth of labour productivity - indicating the presence of substantial increasing returns to scale). The study also finds evidence that technology is diffusing from high to low productivity regions, though the effect is rather weaker. Harris and Law 1998 also find support for the Verdoon effect for UK regions, stating that the coefficient is determined by the combined returns to capital and labour, and arguing that the Verdoon law will be in operation only if there are increasing returns (necessary). Empirical estimare for the UK during 1968-91 show that the presence of increasing returns were greatest in the peripheral regions - possibly due to industry mic e.g. Wales had an industrial mix that resulted in high returns to scale. This is the inherent sectoral bias problem of the verdoon law as industious activities are more susceptible to productivity gains. Boulier 1984 on the other hand claimed that statistical tests have been theoretically unsound and no inferences can be drawn from the empirical results.

OLS assumes that the error terms are not correlated but there is spatial autocorrelation. Regions are fully integrated. Spatial Error Variety: When a shock to labour productivity in one region that is not associated with output growth in that region but will spill-over to also affect labour productivity growth in other regions Spatial Lag Variety: When labour productivity growth in one regions is not only dependent on output growth within that region, but also productivity growth in other regions, whether or not that labour productivity growth is attribuable to output growth. LIKEWISE: OLS assumes output is independent of labout productivity but this assumption can break down due to : 1- Labour -constrained growth as output is endogenous to the employment rate and labour productivity - this however could be rectified by regional labour mobility. 2- Similtaneity bias as in the KDT midel as output has an endogenous element.

Fingleton and McCombie 1998 proposed a solution that states the maximum one commutes on a regular basis is 250 km and so they incorportated this into the Verdoon equation reulting in a high t-ratio for economic growth, thus rejetcung the null hypothesis that Verdoon doesnt hold and finding a signiificantly high spatial interaction which didnt bias the result. To account for omitted variable bias: Fin. and McC 1998 also included a proxy for technological backwardness and found that the Verdoon law still holds.