ورشکستگی و توسعه اقتصادی: تنوع منطقه ای و تنظیم
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|6915||2005||21 صفحه PDF||سفارش دهید||7400 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Economics and Business, Volume 57, Issue 4, July–August 2005, Pages 339–359
We examine the determinants of forced insolvency in New Zealand. The study incorporates three key features. First, we use regional as well as national data. Second, we analyse the role of property prices, which influence collateral values. Third, we explain the rate of total forced insolvency including personal bankruptcies and involuntary company liquidations. Insolvencies are explained by economic activity, financial variables and collateral values. The interactions between economic activity, leverage and property price (collateral) shocks indicate that region-specific shocks can compound into significant localised economic cycles.
The firm life cycle is one where firms are born; many die early; some survive. Of those that survive childhood, some live through to middle age; some successful firms survive with great longevity. Studies at the individual firm level explain factors determining the likelihood of firms dying or progressing from one stage of the life cycle to the next (Bartelsman & Doms, 2000). Another set of studies examines the determinants of the failure rate at the aggregate level, explaining the proportion of firms that fail over time (Wadhwani, 1986; Platt & Platt, 1994; Vlieghe, 2001). This study is in the latter mould, incorporating three key features. The first is that we use regional as well as national data to explain insolvencies. New Zealand insolvency statistics are disaggregated into six regions which have had a variety of economic experiences over the sample period (1988Q1–2003Q2). The start of the period coincides with the mid-point of New Zealand's major economic reforms including removal of agricultural subsidies, reduction of industry protection, labour market deregulation, privatisation, social welfare reform, fiscal consolidation and anti-inflationary monetary policies (Evans, Grimes, Wilkinson, & Teece, 1996). Regional experiences differed markedly through this reform period and in its aftermath. The major urban centres (Auckland, in particular) grew quickly while many peripheral areas either declined or grew only gradually. (Grimes, Kerr, & Aitken, 2003, documents the effects of these developments on regional property prices.) This diversity of experience assists in analysing the determinants of insolvency when the regional data is treated as a panel. Second, we examine the importance of regional property price developments in determining regional insolvency rates. Personal housing is the dominant form of collateral used by small and medium sized enterprises (SMEs) in New Zealand when raising debt finance. Changes in the value of this collateral can influence borrower and lender behaviour and so impact on the rate of insolvency. Third, we explain the total rate of forced insolvency in New Zealand including both personal bankruptcies1 and involuntary company liquidations. We group the two together since many personal bankruptcies are related to small business failures, where the business loan is secured over personal assets (Ministry of Economic Development, 2001). This behaviour is especially relevant to New Zealand which has a large proportion of micro businesses: 86% of all enterprises employ five or fewer full-time equivalent employees (MED, 2003). Our work is informed by a number of other studies, especially Vlieghe (2001) and Platt and Platt (1994). Vlieghe's theoretical approach, which we adapt, is laid out in Section 2. Platt and Platt employ a simple theoretical model but with the distinguishing feature that it is applied as a panel to the states of the USA. They find that they can combine states into four groups; the failure rate of each group is driven by similar variables but with different elasticities. Four variables determine the (log of the) corporate failure rate in their model: the percent change in state employment, percent change in profits earned by sole proprietorships, the (log of the) state real wage and percent change (over 2 years) in the state business formation rate. The two former variables relate to business activity and revenues, the third is a component of business costs and the fourth reflects the higher failure rate for business start-ups than for established businesses. We build on these studies by estimating long run and dynamic models using panel data methods explaining total insolvencies at the regional level. We supplement these estimates with an aggregate national model for comparison. This comparison indicates that the national model fails to detect crucial relationships captured by the regional data. This finding is in keeping with those of Platt and Platt with respect to the United States. In Section 2, we outline our model and the data used for estimation. Section 3 estimates the long run and dynamic models at the aggregate and regional levels. The results are summarised, and their implications discussed, in Section 4. Of particular relevance are the implications of our results for the nature of regional adjustment to national and regional shocks. Also of interest are the implications of our results for the finance-constraint literature which posits that financial variables, possibly unrelated to firm performance, may contribute to firm failure (Greenwald & Stiglitz, 1993). The interaction of the two provides a rich regional dynamic adjustment process in response to regional shocks.
نتیجه گیری انگلیسی
Area-specific shocks impact on incomes and asset values within regions. Our analysis indicates that shocks to regional activity and property values, as well as national developments in inflation and credit provision, impact on the prevalence of regional company failures. Specifically, changes in regional economic activity impact on revenues and thence on the probability of individual company failure, while changes in property prices impact on collateral values. Regional economic activity has a direct effect on property prices (Grimes et al., 2003) but so do other variables (interest rates, new housing supply). Thus, house prices incorporate the effects of shocks over and above the effect of economic activity and act on insolvency rates through a different channel than does economic activity. Faced with a rise in the price of a property used for collateral, a creditor is less pressured to force a debtor into insolvency since the risk to loan repayment is reduced. Conversely, a negative shock to regional property prices results in an increase in the regional level of forced insolvencies. Given that undischarged bankrupts are not allowed by New Zealand law to enter into business alone, be a company director or take part in management of a company (normally for a period of 3 years after initial bankruptcy) a regional property price downturn, with associated increase in insolvencies, may inhibit regional entrepreneurship, at least for a period. Thus, an initial regional downturn can have a magnified and/or prolonged effect on regional economic outcomes.20 An illustration of the interactions between regional economic activity, regional house prices and regional insolvencies is given in Fig. 1 and Fig. 2. We estimate a vector error correction model (VECM) involving these three variables for each of the two smallest regions in our sample: Napier and Dunedin.21 In each case, the VECM incorporates each of the variables included in the equations reported in Table 2 (LTI, LEA, LPR, INFL, LPSC, TIME, seasonals). The first three of these variables are treated as endogenous within the VECM (each being regional variables); INFL and LPSC are treated as being exogenous (being aggregate variables); the non-stochastic variables are also treated as exogenous. The specification includes four lags of the differenced variables plus a single cointegrating vector (CV). The Johansen cointegration test indicates a single CV for Dunedin and two CVs for Napier.22 Therefore, for robustness, we also present results using two CVs in the VECM (Fig. 3 and Fig. 4). The impulse response functions for each endogenous variable in response to one standard deviation innovations in the endogenous variables are shown for a 20-quarter horizon.23 We focus on the response of LEA and LPR to a rise in LTI and the response of LTI to a rise in each of LEA and LPR. A positive innovation to LEA (shown as LEA_N and LEA_D for Napier and Dunedin, respectively) has a permanent (5 years) negative effect on LTI in Napier (with both one and two CVs); the effect in Dunedin is approximately zero (slightly negative with two CVs and slightly positive with one CV). A positive innovation to LPR has at least a transitory negative effect on LTI in each case, with a permanent effect in each of the single CV cases. More clear-cut are the effects of a rise in LTI. A positive innovation to LTI has a permanent negative effect on LEA in each case and a permanent negative effect on LPR. A permanent innovation to LTI may result from a change in legislation pertaining to insolvency. For instance, a change to the legal framework so as to facilitate increased business rehabilitation following a firm's difficulties (in cases where long-term survival of the firm is feasible) could induce a negative innovation to LTI. Such an innovation, according to these simulations, would lead to a (permanent) positive response for regional economic activity and property prices.24 The subsequent interactions of economic activity and property prices (i.e. collateral) with insolvencies helps to sustain the initial beneficial impact of the fall in insolvencies on activity. The VECM results and the collateral effects indicated by our regional equations, help to increase our understanding of the role of financial factors in company insolvency. The estimates indicate that increased aggregate credit provision is a positive contributor to the rate of insolvencies. The New Zealand economy, in keeping with trends in other developed countries, has seen a considerable increase in leverage through the period under consideration.25 This increase in debt has been coupled with a rise in the rate of business start-ups, in part financed by the greater availability of credit. The partial effect of these developments, exhibited through the importance of LPSC in explaining the insolvency rate, is to increase the rate of forced insolvencies. To the extent that increases in real property prices cause real collateral values to keep up with the greater real value of credit, this leverage effect can be mitigated. Reflecting these offsetting effects, there has been no trend increase in aggregate insolvencies in New Zealand since the start of the 1990s despite the strong increase in credit provision. A temporary increase in inflation is another avenue by which the impact of increased leverage on the insolvency rate can be mitigated. As inflation rises, asset values (including collateral and inventory values) rise and firms are better placed, in the short term, to service and/or repay their outstanding debt. However, a longer-term increase in inflation will be reflected in increases in inflation expectations, nominal interest rates and debt servicing, offsetting the beneficial asset value effect. Our estimates over the high inflation period indicate that a rise in inflation loses its beneficial impact on insolvencies under persistent inflationary conditions. The lack of inflation effect on the insolvency rate under these conditions, but significant effect in low inflation conditions, is consistent with the thesis that higher inflation has a short-term positive impact on company balance sheets but only when inflation is unanticipated or otherwise not incorporated into financial prices. Overall, our work reiterates the findings of Platt and Platt, Vlieghe and others that economic activity is an important determinant of insolvencies (both personal bankruptcies and forced company liquidations). It extends their findings by pinpointing the important role also fulfilled by property prices, acting as collateral for loans. Financial market developments have played a major role through the greater provision of credit, contributing to greater leverage and thus greater risk of firms being unable to service and/or repay their debt. Regional developments interact with each of these effects, making regional activity and regional asset values key transmitters of area-specific shocks to regional insolvencies. The feedback of insolvencies to regional economic variables then magnify and prolong regional economic cycles.