Inflation and economic growth nexus in the Southern African Development Community : a panel data investigation

Abstract

English: The aim of the thesis is to examine the relationship between inflation and economic growth using the Southern African Development Community (SADC) as a case study. The motivation emanates not only because of the lack of studies analysing this relationship in the SADC region, but also due to the fact that this relationship may differ from the one that exists in developed countries due to the level of economic development and prudent macroeconomic policies being practised in the latter (Sarel, 1996). The relationship may differ because the vast majority of developed countries have established independent central banks with a clear mandate to keep inflation levels within a specific range (adopted an inflation targeting framework). However, in most developing countries, central banks do not have a clear inflation targeting monetary policy framework, for instance, in the SADC region, only South Africa has adopted an inflation targeting monetary policy framework. High inflation episodes are known to contribute to macroeconomic instability, therefore policy makers find it important to understand the kind of the relationship that exists between inflation and economic growth in order to develop and implement sound macroeconomic policies. Therefore, inflation is viewed to be one of the basic indicators of macroeconomic stability; hence it is an indicator of the ability of the government to manage the economy. High levels of inflation may be indicative of a lack of sound governance by the monetary authority of a country. In addition, it is a sign of government that has lost control of its finances (Fischer,1993). The thesis addresses issues of nonlinearities in the inflation-growth nexus by endogenously estimating the threshold level of inflation below which inflation may have no, or positive, impact on economic growth, or above which inflation may be detrimental to economic growth. It also assesses the effects of a shock to inflation in South Africa, being the largest economy in the region, on inflation and economic growth of the rest of the region. First, different panel data methodologies; Fixed Effects (FE), Difference Generalised Method of Moments (DIF-GMM), System Generalised Method of Moments (SYSGMM), and Seemingly Unrelated Regression (SUR) estimators are used in order to examine the relationship between inflation and economic growth in the region. Second, Panel Smooth Transition Regression (PSTR) methodology is utilised to examine the nonlinearities in the inflation-growth nexus. In particular, the threshold level of inflation is endogenously estimated and the smoothness of the transition from a low to a high inflation regime in the region is also estimated1. Thirdly, the effects of South African inflation on the inflation and economic growth in the rest of the region are assessed using impulse-response functions derived from estimating a Panel Vector Autoregression (PVAR) model. Overall, the study deals with problems which are normally encountered when using cross-country data such as endogeneity, heterogeneity and cross-sectional dependence. The main findings of the study are that inflation and economic growth in the region are negatively related, as is also the case in other regions of the world as depicted by the empirical literature (Fischer, 1993 and De Gregorio, 1993). Therefore, in terms of the inflation-growth link, the SADC region is not different from all the other regions around the globe. Secondly, the threshold level of inflation in the region is estimated at 18.9 per cent, which is in line with the findings of authors like Drukker et al. (2005), Mignon and Villavicencio (2011), and Ibarra and Trupkin (2011), who found a threshold level of 19.2 per cent, 19.6 per cent, and 19.1 per cent for developing countries. However, this threshold level marginally exceeds that of Khan and Senhadji (2001), Schiavo and Vaona (2007), Moshiri and Sepehri (2009) and Espinoza et al. (2010), which studies report threshold values between 10 and 12 per cent for developing countries. The empirical results also reveal that shocks to South African inflation have significant economic impact on inflation, openness, investment and economic growth in the rest of the SADC region. In particular, more interestingly, South African inflation is found to have a negative and statistically significant impact on economic growth in the region for up to about 12 years after the shock, after which, it becomes insignificant. The contribution of the thesis to the literature is that, firstly, this looks into the inflation-growth relationship in the context of Africa, in particular the SADC region; as such an investigation or research has not been conducted before. Secondly, the research takes advantage of panel data methodologies so as to provide more robust estimates and confront the potential bias emanating from problems such as endogeneity, heterogeneity and cross-country dependence that may have affected previous empirical work on inflation-growth nexus. This is believed to provide more informative estimates on the inflation-growth link, and therefore deepens our knowledge of the region.

Sepedi: Maikemišetšo a dinyakišišo tše ke go lekola kamano magareng ga infoleišene le kgolo ya ekonomi ka go šomiša Dinaga tše di Hlabologago tša Borwa bja Afrika (SADC) bjalo ka dinyakišišo ka ga seemo. Lebaka la se ga le tšwe fela go taba ya gore go na le tlhaelelo ya dinyakišišo tšeo di sekasekago kamano ye ka mo seleteng se sa SADC, eupša gape gore ka lebaka la gore kamano ye e ka no fapana le yeo e lego gona ka dinageng tšeo di hlabologilego ka lebaka la maemo a tlhabollo ya ekonomi le melawana ya ekonomi ye kgolo ye e kwagalago ka ge e phethagatšwa ke ekonomi ye kgolo (Sarel, 1996). Kamano ye e ka fapafapana ka lebaka la gore bontši bja dinaga tše di hlabologilego di hlomile dipanka tša bogareng tšeo di ikemego ka mošomo wo o filwego wo o kwešišegago wa go dira gore maemo a infoleišene a be go maemo a itšego (ka go diriša motheo wa go nepiša infoleišene). Le ge go le bjale, ka go dinaga tše dintši tše di hlabologilego, dipanka tša bogareng ga di na le motheo wa melawana wo o kwagalago wa ditšhelete wo o nepišitšego infoleišene, go fa mohlala, ka seleteng sa SADC, ke fela Afrika Borwa ye e dirišago motheo wa melawana ya ditšhelete wo o nepišitšego infoleišene. Ditiragalo tša infoleišene ya godimo di tsebja gore di tsenya letsogo go seemo sa go tekateka sa ekonomi ye kgolo, ka fao bahlami ba melawana ba bona go le bohlokwa go kwešiša mohuta wa kamano ye e lego gona magareng ga infoleišene le kgolo ya ekonomi gore ba kgone go hlama le go phethagatša melawana ye e kwagalago ya ekonomi ye kgolo. Ka fao, infoleišene e bonwa bjalo ka yeo e lego ye nngwe ya dilaetši tša motheo tša seemo sa go se teketeke sa ekonomi ye kgolo; ke ka lebaka leo e lego selaetši sa bokgoni bja mmušo go laola ekonomi. Maemo a godimo a infoleišene a ka laetša gore go na le tlhaelelo ya pušo ye e kwagalago ke ba taolo ya ditšhelete ba ka mo nageng. Godimo ga fao, ke taetšo ya gore mmušo o lahlegetšwe ke taolo ya ditšhelete tša ona (Fischer,1993). Thesese e ahlaahla merero ye ya go se lekalekanele ka seemong sa kgolo ya infoleišene ka go akanya ka gare maemo a mollwane a infoleišene ka fase ga yona go nago le infoleišene ye e ka hlokago, goba ya ba phosethifi, seabe go kgolo ya ekonomi, goba yeo ka godimo ga yona infoleišene e ka šitišago kgolo ya ekonomi. E sekaseka gape diabe tša go phuhlama ga ekonomi ka Afrika Borwa, ke ge e le ekonomi ye kgolokgolo ka mo seleteng, ka ga infoleišene le kgolo ya eknomi ya selete ka moka. Sa mathomo, mekgwa ye e fapanego ya phanele ya tshedimošo; Diabe tša go se Fetoge (FE), Mekgwa ya Ditiragalo tše di Fapanego tša Kakaretšo (DIF-GMM), Mekgwa ya Ditiragalo tša Kakaretšo tša Tshepedišo (SYSGMM), le diakanyi tša Poelomorago ye e sa Amanego ye Nyanyeng (SUR) di a šomišwa ka nepo ya go lekola kamano magareng ga infoleišene le kgolo ya ekonomi ka mo seleteng. Sa bobedi, mokgwa wa Poelomorago ya Phetošetšo go Phanele ye Bonolo (PSTR) o šomišwa go lekola go se lekalekanele ka go dipalopalo tša kgolo ya infoleišene. Kudukudu, maemo a infoleišene a mollwane a akanyetšwa ka gare le bonolo bja phetogelo go tloga go nako ya infoleišene ya fase go ya go ya godimo ka seleteng le yona e a akanywa1. Sa boraro, diabe tša infoleišene ya Afrika Borwa go kgolo ya infoleišene le ya ekonomi ka seleteng ka moka di sekasekwa ka go šomiša mešomo ya phetolo ya kgabišo ye e tšwago go kakanyo ya mokgwa wa Poelomorago ya Bekthara ya Phanele (PVAR). Ka kakaretšo, dinyakišišo di šoma ka mathata ao ka tlwaelo go kopanwago le ona ge go šomišwa tshedimošo ya go ralala le naga ya go swana le tiragalo ya ka gare, ya pharogano le kamanommogo ya makala. Dikutollo tše kgolo tša dinyakišišo ke gore infoleišene le kgolo ya ekonomi ka seleteng di amane gampe, go swana le ka dileteng tše dingwe tša lefase ka ge go laeditšwe ke dingwalwa tša dinyakišišo (Fischer, 1993 le De Gregorio, 1993). Ka fao, mabapi le kamano ya kgolo ya infoleišene, selete sa SADC ga se fapane le dilete tše dingwe go ralala le lefase ka bophara. Sa bobedi, maemo a mollwane a infioleišene ka mo seleteng a akanywa go ba go dipersente tše 18.9, e lego seo se sepelelanago le dikutollo tša bangwadi ba go swana le Drukker et al. (2005), Mignon le Villavicencio (2011), le Ibarra le Trupkin (2011), bao ba hweditšego mollwane wa dipersente tše 19.2, dipersente tše 19.6, le dipersente tše 19.1 tša dinaga tše di hlabologago. Le ge go le bjale, maemo a a mollwane a feta gannyane a Khan le Senhadji (2001), Schiavo le Vaona (2007), Moshiri le Sepehri (2009) le Espinoza et al. (2010), ao a nyakišišago pego ya boleng bja mellwane magareng ga dipersente tše 10 le tše 12 tša dinaga tšeo di hlabologago. Dipoelo tša dinyakišišo le tšona di utolla gore mathata a infoleišene ka Afrika Borwa a na le seabe se segolo sa ekonomi go infoleišene, tlhokegong ya sephiri, peeletšo le kgolo ya ekonomi ka seleteng ka moka sa SADC. Gabotsebotse, le ka fao go kgahlišago, infoleišene ya Afrika Borwa e hwetšwa gore e na le seabe sa go se loke le sa dipalopalo go kgolo ya ekonomi ka mo seleteng go fihla ka mengwaga ye 12 ka morago ga bothata bja phuhlamo, ka morago ga fao, e fetogago go se be bohlokwa. Seabe sa thesese ya dingwalwa ke gore, sa mathomo, se se lebelela kamano ya kgolo ya infoleišene ka seemong sa Afrika, kudukudu ka seleteng sa SADC; ka ge phatišišo goba dinyakišišo tša mohuta woo di se tša dirwa mo nakong ye e fetilego. Sa bobedi, dinyakišišo di šomiša monyetla wa mekgwa ya tshedimošo ya phanele go realo e le go fa dikakanyo tše di tiilego kudu le go lebana lekgonagalo ya go tšea lehlakore fao go bakwago ke kamanommogo ya ka gare, ye e fapanego le ya go ralala le naga yeo e ka bego e amile mošomo wo o fetilego wa dinyakišišo go dipalopalo tša kgolo ya infoleišene. Se go dumelwa gore se fa dikakanyo tša go ba le tshedimošo kudu go kamano ya kgolo ya infoleišene, gomme ka go realo se oketša tsebo ya rena ya selete.

Afrikaans: Die doel van die proefskrif is om die verband tussen inflasie en ekonomiese groei te ondersoek deur die Suider-Afrikaanse Ontwikkelingsgemeenskap (SAOG) as 'n gevallestudie te gebruik. Die motivering spruit nie net voort uit die gebrek aan studies wat hierdie verband in die SAOG-streek ontleed nie, maar ook uit die feit dat hierdie verband kan verskil van die een wat in ontwikkelde lande bestaan ​​as gevolg van die vlak van ekonomiese ontwikkeling en versigtige makro-ekonomiese beleide wat in laasgenoemde toegepas word (Sarel, 1996). Die verband kan verskil omdat die oorgrote meerderheid ontwikkelde lande onafhanklike sentrale banke gestig het met 'n duidelike mandaat om inflasievlakke binne 'n spesifieke reeks te hou ('n inflasieteikenraamwerk aangeneem). In die meeste ontwikkelende lande het sentrale banke egter nie 'n duidelike monetêre beleidsraamwerk wat op inflasie fokus nie. Byvoorbeeld, in die SAOG-streek het slegs Suid-Afrika 'n monetêre beleidsraamwerk wat op inflasie fokus aangeneem. Hoë inflasie-episodes dra by tot makro-ekonomiese onstabiliteit, daarom vind beleidmakers dit belangrik om die soort verhouding tussen inflasie en ekonomiese groei te verstaan ​​om gesonde makro-ekonomiese beleide te ontwikkel en te implementeer. Daarom word inflasie beskou as een van die basiese aanwysers van makro-ekonomiese stabiliteit; daarom is dit 'n aanduiding van die regering se vermoë om die ekonomie te bestuur. Hoë vlakke van inflasie kan dui op 'n gebrek aan gesonde bestuur deur die monetêre owerheid van 'n land. Daarbenewens is dit 'n teken van 'n regering wat beheer oor sy finansies verloor het (Fischer, 1993). Die proefskrif spreek kwessies van nielineariteite in die inflasie-groei-nexus aan deur die drempelvlak van inflasie endogeen te skat, waaronder inflasie geen, of positiewe, impak op ekonomiese groei mag hê nie, of waarbo inflasie nadelig vir ekonomiese groei kan wees. Dit beoordeel ook die gevolge van 'n skok vir inflasie in Suid-Afrika, as die grootste ekonomie in die streek, op inflasie en ekonomiese groei van die res van die streek. Eerstens word verskillende paneeldata-metodologieë gebruik; vaste effekte (VE), verskil-veralgemeende metode van momente (VER-VMM), stelsel-veralgemeende metode van momente (STEL-VMM), en skynbaar onverwante regressie (SOR)-beramers om die verband tussen inflasie en ekonomiese groei in die streek te ondersoek. Tweedens word die paneel-gladde oorgangsregressie (PGOR) metodologie gebruik om die nielineariteite in die inflasie-groei-nexus te ondersoek. In die besonder word die drempelvlak van inflasie endogeen beraam en die gladheid van die oorgang van 'n lae na 'n hoë inflasieregime in die streek word ook beraam. Derdens word die gevolge van Suid-Afrikaanse inflasie op die inflasie en ekonomiese groei in die res van die streek beoordeel deur gebruik te maak van impulsresponsfunksies wat afgelei is van die beraming van 'n paneelvektor-outoregressie (PVOR) model. Oor die algemeen handel die studie oor probleme wat normaalweg teëgekom word wanneer landwyd data gebruik word, soos endogeniteit, heterogeniteit en dwarssnitafhanklikheid. Die hoofbevindinge van die studie is dat inflasie en ekonomiese groei in die streek negatief verwant is, soos ook die geval is in ander streke van die wêreld, soos uitgebeeld deur die empiriese literatuur (Fischer, 1993 en De Gregorio, 1993). Daarom, in terme van die inflasie-groei-skakel, is die SAOG-streek nie anders as al die ander streke regoor die wêreld nie. Tweedens word die drempelvlak van inflasie in die streek op 18,9% geraam, wat ooreenstem met die bevindinge van outeurs soos Drukker et al. (2005), Mignon en Villavicencio (2011), en Ibarra en Trupkin (2011), wat 'n drempelvlak van 19,2%, 19,6% en 19,1% vir ontwikkelende lande gevind het. Hierdie drempelvlak oorskry egter marginaal dié van Khan en Senhadji (2001), Schiavo en Vaona (2007), Moshiri en Sepehri (2009) en Espinoza et al. (2010), watter studies drempelwaardes tussen 10% en 12% vir ontwikkelende lande rapporteer. Die empiriese resultate toon ook dat skokke vir Suid-Afrikaanse inflasie 'n beduidende ekonomiese impak op inflasie, openheid, belegging en ekonomiese groei in die res van die SAOG-streek het. In die besonder, en meer interessant, is gevind dat Suid-Afrikaanse inflasie 'n negatiewe en statisties beduidende impak op ekonomiese groei in die streek het vir tot ongeveer 12 jaar ná die skok, waarna dit onbeduidend word. Die bydrae van die proefskrif tot die literatuur is dat dit eerstens kyk na die inflasie-groei-verhouding in die konteks van Afrika, veral die SAOG-streek, aangesien so 'n ondersoek of navorsing nog nie voorheen gedoen is nie. Tweedens maak die navorsing gebruik van paneeldata-metodologieë om meer robuuste ramings te verskaf en die potensiële vooroordeel wat voortspruit uit probleme soos endogeniteit, heterogeniteit en afhanklikheid tussen lande, wat vorige empiriese werk oor die inflasie-groei-nexus moontlik beïnvloed het, te konfronteer. Daar word geglo dat dit meer insiggewende ramings oor die inflasie-groei-skakel verskaf en dus ons kennis van die streek verdiep.