Back to search

University of Macedonia Economic and Social Sciences

Διεθνής κινητικότητα του κεφαλαίου: το "παρόδοξο Feldstein - Horioka"

Abstract

dc:description

This dissertation studies the so-called 'Feldstein-Horioka paradox'3, according to which there exists a significant, close to unity relation between the magnitudes of gross domestic investment and gross domestic saving of an economy. Feldstein and Horioka (1980)4 found such a close to unity relation for a sample of sixteen developed members of the OECD. Opposite to what common intuition would dictate, the 'FH paradox' as interpreted by FH themselves, seemed to be pointing towards a lack of international capital mobility, in a world thought of as utterly 'globalized', and hence receptive to any kind of goods and capital flows, especially in the era launched by the end of the 1973 oil crisis. Due exactly to its paradoxicality the FH finding triggered a long-lasting and vehement controversy, which has revolved around the implementation of ever more sophisticated econometric techniques, based on the conviction that only this way would a conundrum like the 'FH paradox' finally be resolved. This has also been the reason why we have attempted to structure this dissertation, around the axis of the various econometric methods used so far in the relative literature. In the empirical part of the dissertation, we also adopt many of the aforementioned econometric methods met in the literature, but our main contribution resides in our attempt to better define the variables involved in the initial FH specification from an accounting point of view. More specifically, regarding the saving part (right-hand side) of the classical FH regression equation, we use the balancing item of the 'Use of disposable income account'5 of the 2008 System of National Accounts (SNA) as a proxy for what FH (1980) had called gross domestic saving, deeming it as properly describing the purely domestic constituents of an economy's total saving. Regarding the investment part (left-hand side) of the classical FH regression equation however, we strongly question the extent to which the variable of gross domestic investment which FH (1980) used, really corresponds to an economy's actual gross domestic investment. In fact, FH's gross domestic investment has been the former name of the SNA magnitude of gross capital formation (GCF), which according to the 2008 SNA is ' ... measured by the total value of the gross fixed capital formation6, changes in inventories and acquisitions less disposals of valuables'. Following our electronic correspondence with IMF's specialist on National Accounts and Balance of Payments issues Ms S. Matei7, and given that actually, the majority of the more recent papers within the FH literature have used GFCF as a proxy for FH's variable of 'gross domestic investment', we are also focusing on GFCF in our analysis, instead of the (broader) magnitude of GCF, deeming it more representative of the purely productive part of an economy's domestic investment. However, based on the accounting definitions of both GCF and GFCF one can realize that numerous foreign capital elements might also be included in them8. As a means of 'cleansing' GFCF from any foreign fixed capital assets incorporated in it, we propose instead the magnitude of modified gross fixed capital formation (MGFCF), which is derived by subtracting an economy's imports of fixed capital assets destined to GFCF from the economy's GFCF itself. In the empirical part of the dissertation, we use data for three economically developed countries of Northern Europe, namely Norway, Netherlands and Finland. Using various alternative model specifications, apart from the initial FH one, we find that the generally high values of the saving-investment relation must be attributed to the significant autocorrelation latent in them. Using our own variable MGFCF instead, the aforementioned autocorrelation is significantly reduced or even eliminated. Additionally, we use stationarity and cointegration tests for the variables involved under each one of the various FH model specifications tested. An overall conclusion of our tests, is that the degree of international capital mobility for the three countries of our sample is in fact bigger than the one which the 'FH paradox' would imply. Despite the general lack of data for imports of fixed capital assets from which necessarily suffers the approach we propose, promising perspectives for further research in the future can be discerned.

Degree

thesis:*
Grantor dc:publisher
University of Macedonia Economic and Social Sciences
Year dc:date
2011

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Makedonas, Eleftherios
  • Μακεδόνας, Ελευθέριος

Subjects

dc:subject × 20

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/25378

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Makedonas, Eleftherios; Μακεδόνας, Ελευθέριος. Διεθνής κινητικότητα του κεφαλαίου: το "παρόδοξο Feldstein - Horioka". University of Macedonia Economic and Social Sciences, 2011. http://hdl.handle.net/10442/hedi/25378