Private consumption and sector price behaviour in the Spanish economy: a business cycle approach

Abstract

In this article we examine the consequences of introducing tradable, nontradable and investment goods in a small open economy business cycle model in the case of Spain. The stochastic simulation of the model and its comparison to the Spanish economic data show that the model is able to mimic two prominent features of the Spanish cycle: (1) the dual inflation phenomenon, that is, the cyclical properties of the divergent behaviour between nontradable and tradable prices; and (2) the high volatility of private consumption relative to the volatility of output, a feature known as the Dolado et al. puzzle.

Keywords:

• dual inflation
• business cycle
• volatility
• competitiveness

JEL Classification:

• C63
• C68
• E31
• E32
• O41

Acknowledgements

We acknowledge the financial support from the Spanish Ministry of Science and Innovation and FEDER through grants ECO2008-02752/ECON and ECO2009-10398, the Xunta de Galicia through grant 10PXIB300177PR, Universidad Complutense de Madrid and BSCH and the Fundación Ramón Areces through its grant programme for Research in Economics (Ayudas a la Investigación en Economía).

Notes

¹Germany, Italy and France, basically.

²Wyplosz and Laszlo (1995) and Froot and Kenneth (1994) showed that a higher increase in nontradable prices (versus tradable prices) can be interpreted as a real exchange rate appreciation and this ends up in a competitiveness loss.

³See Balassa (1964) and Samuelson (1964).

⁴See Dolado et al. (1993).

⁵Theories like life-cycle (Modigliani, 1966) or permanent income (Friedman, 1956) suggest that households smooth consumption along the business cycle. Hence, volatility of consumption relative to volatility of output should be low.

⁶See Puch and Licandro (1997).

⁷In the case of Spanish economy, the nontradable share is 70.3% versus a share of around 50% in Canada, France, Italy, the United Kingdom and the United States, for example.

⁸See Katsimi (2004) for econometric analysis of the Balassa–Samuelson effect applied to Euro area.

⁹The investment good is produced using traded and nontraded inputs, which can be thought of loosely as equipment and structures, respectively.

¹⁰Kehoe and Fernández de Córdoba (2000) introduced this type of production function for the investment sector.

¹¹Computing business cycle dynamics in the standard small open economy model is problematic because the dynamics of the model implies the unconditional variance of variables such as holdings and consumption is infinite. Then we adopt this specification to induce stationarity of equilibrium dynamics.

¹²This formulation for the risk premium follows Lim and McNelis (2008).

¹³See Prescott (1986).

¹⁴It would have been preferable to subtract electricity, gas and water out of the traded category and to add transportation services. Unfortunately, there are not available data in the Spanish National Accounts.

¹⁵We exclude nonmarket services from the analyses because they are not traded on a market and therefore their output and prices can be considered accounting conventions.

¹⁶This classification is used also by Salido and Pérez Quirós (2005).

¹⁷This is a feature also present in other papers applied to the Spanish economy with different temporal horizons. See, for example, the paper of Puch and Licandro (1997) and Giménez and Martín-Moreno (2002).