A. J. Hughes Hallett

Does trade integration cause convergence

Abstract Using a business cycle model, we analyse the conditions for trade integration to produce economic convergence. We find large, stable economies with integrated structures are likely to diverge. But smaller, more volatile or less well integrated economies will converge. The symmetry of shocks is unimportant.

Testing for hysteresis against nonlinear alternatives

Abstract A number of studies have hypothesised certain economic behaviour to be hysteretic, but few have actually been able to establish hysteretic behaviour in the data. This is because of the lack of an effective test for hysteresis. This paper compares and contrasts two new measures of hysteretic behaviour, designed explicitly for detecting hysteresis in economic systems. Using Monte Carlo techniques, we show that one has considerably more power than the other against nonlinear, cyclical or multiple equilibria alternatives in the presence of random disturbances. The key properties of this measure appear to be functional form independence and rate independence.

Unification and the policy predicament in Germany

We argue that wages have increased so far ahead of labour productivity in East Germany as to produce a problem that will continue to hound German policy-makers for the next two decades. Despite rapid rates of capital accumulation (around 9%) and growth (around 5%) in East Germany over the coming ten years, our estimates show that even if wage catch-up decelerates greatly, as long as it continues, the rate of unemployment in the East will still be twice as high as in the West in another ten years. Alternatively, if wage discipline forces the Eastern unemployment rate to come down to the Western level, wage differentials will widen substantially over these next ten years. Thus serious problems loom ahead.

AGGREGATE PHILLIPS CURVES ARE NOT ALWAYS VERTICAL: HETEROGENEITY AND MISMATCH IN MULTIREGION OR MULTISECTOR ECONOMIES

The aggregation of sectoral or regional Phillips curves yields an inflation–unemployment trade-off that is not vertical in the long run if there are mismatches between supply and demand in the regional or sectoral labor markets. This remains true even when the individual Phillips curves are all vertical. This result stems from variations in the slope of the individual short-run Phillips curves, rather than from changes to the equilibrium level of unemployment. It implies a role for the management of the distribution of demand over different sectors or regions, in order to minimize the natural rate of unemployment.

Nonlinearity, ComputationaL Complexity and Macroeconomic Modelling

In this paper we survey and appraise the main contributions to solving and stabilising non-linear equation systems typically found in Economics. We are keen wherever possible to draw distinctions and limiting cases between different solution methods, define acceleration strategies and encourage the use of hybrid or switching algorithms. Both large-scale traditional macroeconomic models as well as smaller non-linear analytical models are considered. Copyright 1999 by Blackwell Publishers Ltd

Hybrid algorithms with automatic switching for solving nonlinear equation systems

Abstract An optimal switching device is introduced for hybrid algorithms which try to combine the superlinear convergence speeds of Newton model solution techniques with the cheapness of slower Gauss-Seidel techniques. We also propose a device for combining second-order and Gauss-Seidel iterative techniques. Tests show that these hybrids lead to powerful improvements in solving nonlinear models, both in terms of accelerating convergence and reducing computational costs and in extending the radius of convergence in difficult cases.

Real adjustment in a union of incompletely converged economies: An example from East and West Germany

Abstract Economic and monetary reunification in Germany has proved to be more expensive than previously thought - and not just for the Germans. If a ‘Mezzogiorno problem’ of continuing fiscal transfers to the East, and possible migration flows west, are to be avoided, then there has to be convergence in productivity levels. This paper analyses possible convergence paths, and the policy regimes which accelerate convergence. The intention is to illustrate the problems facing a European monetary union of asymmetric and incompletely converged economies. We find that convergence is likely to be slow: perhaps 30–40 years in the German case, despite very fast growth in the East. Second, a very substantial part of the servicing and subsidising costs have to be paid by other (non-German) economies in the union, without any obvious compensating benefits. Third, to reduce the need for continuing transfers actually requires a policy which promotes price and wage flexibility in the depressed region. That appears to run counter to the current market integration within Europe.

On the Asymmetric Impacts of a Common Monetary Policy in Europe

Economists have long feared that economic convergence towards monetary union might be superficial or short-lived; more the product of political will than of economic advantage or market incentives. The worry is the convergence criteria were defined in nominal terms, and for one particular moment of time. This says nothing about convergence in structures and responses, or about the ability to remain converged over a period of time. If significant differences remain in structures, or in the national responses to policy changes, shocks, or other events, then it is inevitable that common policies — and a common monetary policy in particular — will have different impacts in different places. That could delay convergence, if not start to drive the union’s economies apart.

Policy bargains and the problem of model selection

Abstract Parameter uncertainty and disagreements about model specification pose a particular difficulty for economic policy design. Frankel and Rockett argue that, if model uncertainty is not explicitly taken into account, policy coordination is as likely to decrease welfare as increase it. This paper therefore looks at the scope for reducing errors by making strategic model choices part of the overall policy selection procedure. It shows that alternative definitions of the policy bargain produces decisions which are robust against disagreements about the correct model; while alternative choices of model produce policies which are robust against model uncertainty. Hence fixed (i.e., model-free) policy rules are no substitute for making strategic model choices.

On the Accuracy and Efficiency of GMM Estimators: A Monte Carlo Study

GMM estimators are now widely used in econometric and financial analysis. Their asymptotic properties are well known, but we have little knowledge of their small sample properties or their rate of convergence to their limiting distribution. This paper reports small sample Monte Carlo evidence which helps discriminate between the many GMM estimators proposed in the literature. We add a new GMM estimator which delivers better finite sample properties. We also test whether biases in the parameter estimates are either significant or significantly different between estimators. We conclude that they are, with both relative and absolute biases depending on sample size, fitting criterion, non-normality of disturbances, and parameter size.