STRUCTURAL BREAKS AND UNIT ROOTS IN AUSTRALIAN MACROECONOMIC TIME SERIES

Abstract.  We examine the unit root properties of 16 Australian macroeconomic time series using monthly data spanning the period 1960–2004. In addition to the standard Augmented Dickey Fuller (ADF) test, we implement one- and two-break endogenous structural break ADF-type unit root tests as well as one- and two-break Lagrange multiplier (LM) unit root tests. While the ADF test provides relatively little evidence against the unit root null hypothesis, once we allow for structural breaks we are able to reject the unit root null for just under half of the variables at the 10% level or better.     

A reexamination of trends in primary commodity prices

This paper conducts a reexamination of the Prebisch-Singer hypothesis by employing the unit root test proposed by Lee and Strazicich (2003) and Lee and Strazicich (2004) that allow for up to two structural breaks. Given the higher power of these tests compared to the Zivot and Andrews (1992) and Lumsdaine and Papell (1997) tests, rejection of the null can be considered as genuine evidence of stationarity. The main findings of this paper are that eleven out of twenty-four commodity prices are found to be difference stationary implying that shocks to these commodities tend to be permanent in nature. The remaining thirteen prices are found to exhibit trend stationary behavior with either one or two structural breaks. Most of the commodities that do not exhibit difference stationary behavior seem to contain no significant trends. There are fewer cases, in relation to past studies, of commodities that display negative trends thereby weakening the case for the Prebisch-Singer hypothesis.     

Are business cycles stationary fluctuations around a deterministic trend? Empirical evidence from 79 developing countries

There is a large literature that tests the univariate time series properties of the real output series following the seminal work of Nelson and Plosser (1982). Whether or not real output is characterized by a unit root process has important implications. A unit root in real output, for instance, is inconsistent with the notion that business cycles are stationary fluctuations around a deterministic trend. In this paper, we investigate the univariate time series properties of real output for 79 developing countries using the conventional augmented Dickey and Fuller (1979) unit root test, the Zivot and Andrews’ (1992) one structural break unit root test, and the Lumsdaine and Papell (1997) two structural breaks unit root test. Our main finding is that, for 40 countries, real output is stationary around a trend. This indicates that business cycles are stationary fluctuations around a deterministic trend for only 51% of the developing countries in our sample.     

Alternative lag length selection criteria and the split-trend stationarity hypothesis

This study asks whether evidence that key macroeconomic time series are stationary around broken trends is robust to using different criteria to determine the lag length in the ADF regressions. When lag lengths are determined using the Schwarz criterion or two different specific-to-general methods, tests for unit roots in several series in the Nelson-Plosser (1982) data and in US postwar real GNP find weaker evidence against the unit root hypothesis than either Perron (1989), who set the date for the break in the trend a priori or Zivot and Andrews (1992), who determined the break date endogenously.     

‘I’ve been to Bali too’ (and I will be going back): are terrorist shocks to Bali's tourist arrivals permanent or transitory?

International visitor arrivals to Bali are examined using univariate and panel Lagrange multiplier (LM) unit root tests with one and two structural breaks to ascertain if shocks to the time path of tourist arrivals are permanent or transitory. The univariate LM unit root tests with one and two structural breaks fail to reject the null hypothesis of a unit root in international visitor arrivals to Bali. However, the panel LM unit root tests with one and two structural breaks applied to a panel of Bali's 11 major source markets reject the null and support the alternative hypothesis of a joint trend-stationary series with transitory shocks. This result suggests that, the effects of the recent terrorist acts on Bali on the growth path of tourist arrivals from major markets are only transitory and that as a consequence Bali's tourism sector is sustainable in the long run.     

A nonlinear approach to testing the unit root null hypothesis: an application to international health expenditures

In this article, we examine the unit root null hypothesis for per capita total Health Expenditures (HEs), per capita private HEs and per capita public HEs for 29 Organization for Economic Co-operation and Development (OECD) countries. The novelty of our work is that we use a new nonlinear unit root test that allows for one structural break in the data series. We find that for around 45% of the countries, we are able to reject the unit root hypothesis for each of the three HE series. Moreover, using Monte Carlo simulations, we show that our proposed unit root model has better size and power properties than the widely used Augmented Dickey–Fuller (ADF) and Lagrange Multiplier (LM) type tests.     

Are US macroeconomic series difference stationary or trend-break stationary?

This article tests for the presence of a unit-root in all time series included in the extended Nelson–Plosser data set using the statistics devised by Zivot and Andrews, Perron and Murray and Zivot. It specifies the mixed model characterization of the trend-break stationary alternative that allows for a simultaneous break in both the intercept and slope of the trend-function. It rejects the unit-root null hypothesis for real GNP, nominal GNP, real per capita GNP, industrial production, employment, GNP deflator, nominal wages, interest rate and common stock prices. Use of appropriate critical values to assess the significance of the trend-function coefficients reveals that the slope-break should be included in real GNP, nominal GNP, real per capita GNP, nominal wages, interest rate and common stock prices. The results indicate that there is less evidence against the unit-root hypothesis with the extended Nelson–Plosser data compared to the original Nelson–Plosser data.     

Removal of farm subsidies in a regional economy: a computable general equilibrium analysis

There are a number of studies that examine the purchasing power parity (PPP) hypothesis. The empirical findings from the extant literature for the PPP hypothesis are mixed. This article applies univariate and panel Lagrange Multiplier (LM) unit root tests with one and two structural breaks to real exchange rates for 15 Asian countries. The univariate LM unit root tests find evidence of PPP for two-thirds of the sample. The results from the panel LM unit root test support long-run PPP for the Asian countries in the sample. The results from the LM panel unit root tests differ from those of existing panel unit root tests of PPP for Asian countries that have not allowed for the existence of structural breaks.     

A pair-wise approach to output convergence between European regions

We apply the Pesaran (2007) pair-wise approach of convergence to the per capita outputs of 195 European regions for the period 1980-2006. Pesaran's approach is based on the computation of the percentage ratio of output gaps which fulfil a given convergence criterion. A high ratio will be interpreted in favour of convergence. In a first step, we define stochastic convergence between two regions as level stationarity of their output gap. Deviations from its equilibrium value will only have a temporary effect. Results from several usual unit root or stationarity tests show us that the percentage ratio of level stationary output gaps is low, which stands against this definition of convergence. However, this convergence criterion excludes the possibility of changes in output gap equilibrium value or catching up between regions. To fit these cases, we combine the pair-wise approach with unit root or stationarity tests with structural breaks. Structural breaks are modelled by dummies (Zivot and Andrews, 1992; Kurozumi,2002) or as smooth structural breaks (Christopoulos and León-Ledesma, 2009). Overall results are not changed as convergence is not accepted more often. Finally, we consider the autocorrelation function approach of Caggiano and Leonida (2009). Autocorrelations and their confidence intervals are estimated for each output gap. Convergence between two regions is accepted if their per capita output gap autocorrelations become nonsignificantly different from zero after some lag. Results show that a high percentage of regions satisfy this convergence criterion. Contrary to the conclusions which could be made from previous results, shocks to output gaps seem to disappear as time passes.     

ARE LABOUR FORCE PARTICIPATION RATES NON‐STATIONARY? EVIDENCE FROM 130 YEARS FOR G7 COUNTRIES*

This paper applies a unit root test with a non‐linear threshold to examine whether labour force participation rates are mean reverting for G7 countries using annual data over a 130 year period. We find some evidence of mean reversion for just over half the sample; however, this result is sensitive to regime shifts. We also examine whether the labour force participation rate is trend reverting through employing a lagrange multiplier (LM) unit root test with one and two structural breaks in the intercept and slope. The LM unit root test provides no additional evidence in support of stationarity. On the basis of the unit root tests for mean reversion we conclude that there is at best mixed evidence that long‐term changes in unemployment rates translate into long‐term changes in employment rates and that the unemployment rate is a useful indicator of joblessness.     

Structural break,nonlinearity and asymmetry: a re-examination of PPP proposition

In this study, we examine the validity of the PPP proposition for 28 European countries. For this purpose, we propose a new unit root test procedure that allows for both gradual structural breaks and asymmetric nonlinear adjustment towards the equilibrium level. Small-sample properties of the new tests are examined through Monte-Carlo simulations. The simulation results suggest that the new tests have satisfactory size and power properties. We then apply these new tests along with other unit root tests to examine stationarity properties of real exchange rate series of the sample countries. Our tests reject the null of unit root in more cases when compared to alternative tests. Overall, we find that the PPP proposition holds in majority of the European countries examined in this article.     

The Feldstein–Horioka Puzzle and structural breaks: Evidence from EU members

The purpose of this paper is to investigate the level of capital mobility in European Union members using the Feldstein–Horioka puzzle proposed by Feldstein and Horioka (1980) in order to investigate relations between saving and investment flows. In this paper, data for 23 European countries were used over the period of 1995–2009 on the quarterly basis. Two different tests were used to estimate the stationarity of the model variables, which are the Ng and Perron (2001) unit root test procedure and approach proposed by Zivot and Andrews (1992) for unit root test allowing for a structural shift. Then the Kejriwal and Perron (2008, 2010) structural break test was applied to determine the presence of structural breaks in series. In most countries except Belgium and Finland UDmax and WDmax tests rejected the hypothesis of no breaks. To test the cointegration relationships between investment and saving flows of European Union members three different cointegration techniques were applied to the data. Firstly, the Johansen (1988) cointegration approach was used for the case of no cointegration shifts, then the Gregory and Hansen (1996) cointegration test was applied, which allows for one structural shift. Finally, again the Johansen' cointegration approach was used; however, this time with the inclusion of dummy variables related to earlier selected structural break locations. The empirical results provided stronger evidence of cointegration between investment and saving variables in the case of structural break accommodation compared to the case where the presence of structural breaks was ignored. The estimated saving retention coefficient in the presence of structural breaks using the Kejriwal and Perron (2008, 2010) approach appeared relatively low in many cases, illustrating by this the openness of estimated countries. In general, world and European countries with time have a tendency to a higher level of their capital market openness. Estimations of a saving retention coefficient in the presence of structural changes do not support the existence of the Feldstein–Horioka Puzzle in the considered EU countries, except Belgium.     

Fundamentals and Oil Price Behaviour: New Evidence from Co‐integration Tests with Structural Breaks and Granger Causality Tests

As the world economy has undergone various changes and shocks, the oil market went through significant fluctuations during the period 1994–2015. This study focuses on discussing the possible factors that determine crude oil prices, which include world economic growth, market power of Organization of the Petroleum Exporting Countries (OPEC), non‐OPEC supply and the value of dollar, taking into account the structural changes that influence the study period, which is quarterly data for the period of 1994.Q1–2015.Q3. For time series stationarity tests, Zivot and Andrews (1992) and Perron (1997) unit root test with structural break is used. To test the existence of a long relationship in the presence of structural changes, the Gregory and Hansen (1996a, 1996b) method of co‐integration is used. For long‐run coefficient, we proceed to estimate fully modified least squares. The result shows a significant influence of non‐OPEC supply, the dollar appreciation and world gross domestic product growth, but OPEC did not have a significant effect on the price of oil, which is indicated by the structural break for OPEC capacity utilisation at 2002.Q1, that indicates the starting point of the loss of OPEC power. Establishing the presence of co‐integration, we apply the evaluation of Granger causality for co‐integrated data, using vector error‐correction model.     

Do Malaysian house prices follow a random walk? Evidence from univariate and panel LM unit root tests with one and two structural breaks

In this article we apply univariate and panel Lagrange Multiplier (LM) unit root tests with one and twostructural breaks proposed by Lee and Strazicich (2003, 2004) and Im et al. (2005) to examine housing prices for five different housing price indices (all housing, detached housing, semi-detached housing, terrace housing and high-rise housing) in 14 states of Malaysia to test whether housing prices exhibit a random walk. Our main finding from the univariate LM unit root tests is that for the vast majority of states housing prices follow a stationary process about a segmented trend. The results of the panel LM unit root tests provide overwhelming evidence that house prices are segmented trend reverting.     

On the power of bootstrap tests for stationarity: a Monte Carlo comparison

This paper studies the behavior of recently proposed bootstrap tests for the null hypothesis of stationarity when the data are generated under the alternative hypothesis of a unit root. Using Monte Carlo experiments and empirical examples, it is shown that the power of these tests critically depends on the type of bootstrap employed. Specifically, while tests based on the stationary bootstrap have power functions that are increasing with respect to sample size, those based on the sieve bootstrap have non-monotonic power functions. We argue that this difference arises from the fact that the latter procedure does not impose the null hypothesis when generating the bootstrap samples while the former ensures that the bootstrap samples are stationary, conditional on the original data. Our results therefore suggest that while both forms of bootstrap are effective at providing improved distributional approximations under the null hypothesis, it is important to pay careful attention to the particular type of bootstrap being employed when attempting to distinguish between the unit root and stationarity hypotheses as the choice of bootstrap can have crucial implications for the power of the resulting tests.     

The structure of tourist expenditure in Fiji: evidence from unit root structural break tests

The central aim of this paper is to investigate whether shocks to Fiji's tourism industry have a permanent effect or a transitory effect on tourist expenditure in Fiji. To accomplish this aim the Zivot and Andrews (1992 Zivot, E and Andrews, D. 1992. Further evidence of the great crash, the oil-price shock and the unit-root hypothesis. Journal of Business and Economic Statistics, 10: 251–70.  ) one break test and the Lumsdaine and Papell (1997 Lumsdaine, R and Papell, D. 1997. Multiple trend breaks and the unit root hypothesis. Review of Economics and Statistics, 79: 212–18.  ) two break tests are used. The one break and two break tests reveal 1987 – the year of the military coups in Fiji – as the year of the break. Moreover, it is possible to reject the unit root null leading to the conclusion that shocks to Fiji's tourism industry have a transitory effect on tourist expenditure in Fiji.     

DOES DISAGGREGATION AFFECT THE RELATIONSHIP BETWEEN HEALTH CARE EXPENDITURE AND GDP? AN ANALYSIS USING REGIME SHIFTS*

This paper investigates the impact of policy shifts on disaggregated health expenditure‐GDP relationship in Australia. In contrast to previous studies the disaggregation is at the level of type of service delivered and not at the level of source of expenditure. Our results show that the subcomponents of health expenditure exhibit different patterns of behaviour at both cointegration and unit root stages once policy shifts or structural breaks are allowed in the empirical analysis. When the possibility of structural break is allowed we find a significant long‐run relationship between subcomponents of aggregate health expenditure and GDP that is not observed when break is not allowed. The underlying reasons for the occurrence of breaks and policy lessons are discussed subsequently.     

Do per capita health care expenditures converge among OECD countries? Evidence from unit root tests with level and trend-shifts

This study examines the stochastic conditional convergence of per capita health care expenditures (PCHCE) among 19 OECD countries over the period 1972–2008. Specifically, newly developed LM and RALS-LM unit root tests with allowance for two endogenously determined structural breaks are employed. The results indicate support for PCHCE convergence among most OECD countries. The results are stronger in more general tests that control for two breaks and nonnormal errors. Panel unit root tests provide additional support for the stochastic convergence of PCHCE.     

Quantile unit root test and the PPP in Africa

In this study, we apply the Quantile unit root test and revisit the Purchasing Power Parity (PPP) in 20 African countries using real effective exchange rates over the period 1971Q1 to 2012Q4. While traditional unit root tests fail to reject unit root hypothesis in most of the countries, results from Quantile unit root test reject unit root null hypothesis in Ghana, Mauritius, Niger, South Africa, and Togo, providing support for the PPP at least in these five countries. We further estimate the half-life based on Quantile autoregressive (QAR) model to be about 4.57–7.96 quarters (1–2 year).     

Persistence in Corporate Performance? Empirical Evidence from Panel Unit Root Tests

Persistence in corporate performance is analyzed in the framework of empirical tests of unit root behavior concerning firm profits. Data for firm-specific rates of return is applied in a set of panel unit root tests to address the question of persistence in profits both at firm level and for the aggregate level of industry-specific profits. The firm data all reject a null hypothesis of random walk behavior of profits but when smoothing profit rates at a two-digit NACE-code level for industries, the empirical evidence is more mixed as most industries show up with a unit root in aggregate rates of return, i.e. indicating persistence in corporate performance.