.
Annunci online

 
La matematica delle rivoluzioni
post pubblicato in diario, il 30 agosto 2011



 
Osservando questa tabella si vede che guerre, moti, manifestazioni e/o rivoluzioni, avvengono dal 1600 in poi con cadenza circa 20ennale. 20 anni corrispondono per la specie umana all'incirca all'età in cui viene riprodotta una nuova generazione. Nel 1600 le guerre erano più frequenti perché i regnanti avevano la necessità di far sfogare l'aumento demografico in tale modo. Quando non ci sono state guerre si sono verificati moti o rivoluzioni. Nella tabella ci sono dei periodi di latenza dovuti alla durata della guerra precedente di cui ho scritto solo l'inizio, eventuali pestilenze o epidemie (tipo Spagnola del 1918) che spiegano eventuali buchi della teoria.
-----------------------------------------

 

Tabella che mette in relazione il numero della popolazione umana quando raggiunge uno dei rispettivi numeri di Fibonacci, con l'anno e l'innovazione corrispondente.

-----------------

Ponticelli e Voth dell'università di Barcellona hanno fatto uno studio matematico, scientifico e oggettivo di come avvengono le rivolte nei paesi occidentali. Se Hegel e i filosofi idealisti storicisti avessero avuto questi dati, avrebbero fatto salti di gioia!

 

Traduzione dell' Abstract di Fabio Marinelli. Mi sto cimentando nella traduzione completa.

Al seguente link si trova il documento pdf originale:

 http://www.voxeu.org/sites/default/files/file/DP8513.pdf

 

ISSN0265-8003
 
 
AUSTERITY AND ANARCHY: BUDGET CUTS AND SOCIAL UNREST IN EUROPE, 1919-2009
 
 
Jacopo Ponticelli, Universitat Pompeu Fabra
Hans-Joachim Voth, UPF-ICREA, CREI and CEPR
 
 
Discussion PaperNo. 8513
August 2011
 
 
Centre for Economic Policy Research
77 Bastwick Street, London EC1V3PZ, UK Tel: (44 20) 71838801, Fax: (44 20) 7183 8820
 
 
 
 
This Discussion Paper is issued under the auspices of the Centre’s research programme in                       INTERNATIONAL          MACROECONOMICS.    Any    opinions expressed here are those of the author(s) and not those of the Centre for Economic  Policy Research. Research disseminated by CEPR may include views on policy, but the Centre itself takes no institutional policy positions.
 
 
The Centre for Economic Policy Research was established in 1983 as an educational  charity, to promote independent analysis and public discussion of open  economies and the relations among them. It is pluralist and non-partisan, bringing economic research to bear on the analysis of medium- and long-run policy questions.
 
 
These Discussion Papers often represent preliminary or incomplete work, circulated to encourage discussion and comment. Citation and use of such a paper should take account of its provisional character.
 
 
Copyright: Jacopo Ponticelli and Hans-Joachim Voth
  
ABSTRACT
  
Austerity and Anarchy: Budget Cuts and Social Unrest in Europe,
1919-2009*
 
 
Does fiscal consolidation lead to social unrest? From the end of the Weimar Republic  in Germany  in  the  1930s  to  anti-government  demonstrations  in Greece in 2010-11, austerity has  tended to go hand in hand with politically motivated violence and social instability. In this paper,  we  assemble cross-country evidence for the period 1919 to the present, and examine the extent to which societies become unstable after budget cuts. The results show a clear positive correlation between fiscal retrenchment and instability. We test if the relationship simply reflects economic downturns, and conclude that this is not the  key  factor.  We  also  analyse  interactions  with  various  economic  and political variables. While autocracies and democracies show a broadly similar responses to budget cuts, countries with more constraints on the executive are less likely to see unrest as a result of austerity measures. Growing media penetration does not lead to a stronger  effect of cut-backs on the level of unrest.
 
 
 
 
 
Il risanamento delle finanze pubbliche comporta tensioni sociali?  Dalla fine della Repubblica di Weimar in Germania nel 1930 alle manifestazioni anti-governative in Grecia nel 2010-11, l'austerità ha teso a procedere di pari passo con la violenza di matrice politica e instabilità sociale. In questo lavoro, abbiamo assemblato le prove attraverso i paesi, per il periodo dal 1919 ad oggi, per esaminare la misura in cui le società diventano instabili dopo i tagli di bilancio. I risultati mostrano una chiara correlazione positiva tra i tagli fiscali e l’instabilità. Testiamo se il rapporto riflette semplicemente recessioni economiche, per concludere che questo non è il fattore chiave. Noi abbiamo anche analizzato le interazioni con le diverse componenti economiche e le variabili politiche. Mentre autocrazie e democrazie mostrano una risposta sostanzialmente simile quando ci sono i tagli di bilancio, i Paesi con più vincoli sull’esecutivo hanno meno probabilità di subire disordini come risultato di misure di austerità. 
La crescente penetrazione dei media non porta ad un effetto più forte dei tagli sul livello dei disordini.     
                                                                  

1. Introduction

Social unrest  has led to key turning points  in modern  history  since, at  least, the French Revolution. Marx saw it as the  driving force of the  transition of societies from feudalism  to capitalism and,  eventually, communism. Unrest’s power as a catalyst for change  manifests itself explicitly  regime changes,  such as during  the  “Arab  Spring” of 2010-2011, or it operates through expectations: The  extension  of the  franchise  in Western societies has been interpreted as an attempt to heed off the  threat of revolution  (Acemoglu  and  Robinson  2000).1

 

What leads to social unrest  is less clear. Economic  shocks are one important contributing factor:  The  demise of the  Weimar  Republic  during  the  Great  Depression is a prominent example  of how economic hardship  can translate into  unrest  (Bracher 1978).2
In this  paper,  we examine  what  leads to social instability and  violent protests. In particular, we ask whether  fiscal policy affect the  level of social unrest.  The  extent  to which societies fracture  and  become unstable in response to drastic  changes  in the government budget is a primary  concern  for policy makers  attempting to reduce budget deficits:  From Argentina  in 2001 to Greece in 2010-11, austerity measures  have often created  a wave of violent protests and massive civil unrest.  Economic  conditions  can deteriorate further  and faster  if political  and social chaos follows attempts to reign in spending. Consequently, sustainable debt  levels for countries  that are prone  to unrest may be lower than  they  otherwise  would be.
We use a long panel dataset covering almost  a century, focusing on Europe,  1919 to 2009. The  continent went  from high levels of instability in the first half of the  20th century  to relatively  low ones in the  second, and  from frequently  troubled economic conditions  to prosperity. It thus  provides  a rich laboratory of changing  economic, social and  political conditions.  In terms  of outcome  variables,  we focus on riots, demonstrations, political assassinations, government crises, and attempted revolutions. These  span  the  full range  of forms of unrest,  from relatively  minor  disturbances to armed  attempts to overthrow  the established political  order.  We compile a new index that summarizes  these  variables,  and  then ask  -- for every percentage cut  in government spending,  how much  more instability should  we expect?
The data  shows a clear link between  the  magnitude of expenditure cut-backs and increases  in social unrest.  With  every additional percentage point  of  GDP  in spending cuts,  the  risk of unrest  increases.  As a first pass at the  data, Figure 1 examines  the relationship between  fiscal adjustment episodes and  the number of incidents indicating instability (CHAOS).  CHAOS  is the sum of demonstrations, riots,  strikes, assassinations, and  attempted revolutions  in a single year in each country. The  first set of five bars show the  frequencies condition alon the  size of budget  cuts. When  expenditure is increasing,  the average  country-year unit  of observation in our data  registers  less than  1.5 events.  When expenditure cuts  reach  1% or more of GDP,  this  grows to nearly 2 events,  a relative  increase by almost  a third  compared  to the  periods of budget expansion.  As cuts  intensify,  the  frequency  of disturbances rises. Once austerity measures involve expenditure reductions by 5% or more, there are more than  3 events per year and country-- twice as many  as in times of expenditure increases.
 
1  In a related exercise, Boix (2003) models the incentives of the populace to resort to violence as a function of the wealth distribution and economic development.
2  The French Revolution has also been interpreted in these terms (Soboul 1974; Doyle 2001). The view is controversial (Hunt 2004; Cobban 1964).

 


1) Introduzione (In verde: traduzione in italiano di Fabio Marinelli).

I Disordini sociali hanno portato a punti di svolta nella storia moderna almeno per  la Rivoluzione francese. Marx la vedeva come la forza motrice del passaggio delle società dal feudalesimo al capitalismo e, infine, al comunismo. La potenza dei disordini si manifesta esplicitamente come un catalizzatore per il cambiamento di regime,come ad esempio durante la "primavera araba" del 2010-2011, oppure opera attraverso le aspettative: L'estensione del diritto di voto nelle società occidentali è stato interpretato come un tentativo di ascolto per prevenire la minaccia dirivoluzione (Acemoglu e Robinson 2000) 1. Cosa porta a tensioni sociali è meno chiaro. Le crisie conomiche sono un importante fattore contribuente: La fine della Repubblica di Weimar durante la Grande Depressione è un esempio lampante di come la crisi economica può tradursi in disordini (Bracher 1978) 2. In questo articolo, esaminiamo ciò che porta a instabilità sociale e violente proteste. In particolare, ci chiediamo se la politica fiscale influenza il livello della tensione sociale. La misura in cui le società si frantumano e diventano instabili in risposta ai drastici cambiamenti nel bilancio del governo è una preoccupazione primaria per i politici che tentano di ridurre il deficit di bilancio: dall’ Argentina nel 2001 alla Grecia nel 2010-11, le misure di austerità hanno spesso creato un'ondata di violente proteste e disordini di massa. Le condizioni economiche possono peggiorare ulteriormente e più velocemente se il caos politico e sociale segue i tentativi di ridurre la spesa. Di conseguenza, la soglia di debito sostenibile per i paesi che sono inclini a disordini può essere più bassa di quanto sarebbe altrimenti. Noi utilizziamo un lungo dataset  che copre quasi un secolo, concentrandosi sull’ Europa, dal 1919 al 2009. Il continente è passato da alti livelli di instabilità nella prima metà del 20°secolo a quelli relativamente bassi della seconda metà, e da condizioni spesso difficili per la prosperità economica. Esso fornisce quindi un laboratorio ricco di cambiamenti delle condizioni economiche, sociali e politiche. Intermini di variabili di risultato, ci concentriamo su sommosse, dimostrazioni, omicidi politici, crisi di governo, e tentate rivoluzioni. Questi coprono l'intera gamma di forme di agitazione, da disturbi relativamente minori ai tentativi armati di rovesciare l'ordine politico. Si compila un nuovo indice che riassume queste variabili, per poi chiedere: “per ogni taglio percentuale della spesa pubblica, quanta maggiore instabilità ci dobbiamo aspettare?" I dati mostrano un chiaro legame tra l'entità della spesa dei tagli e aumenti di disordini sociali. Con ogni punto percentuale di PIL di ulteriore taglio della spesa, aumenta il rischio di disordini. Come primo passo dei dati in figura 1 si esamina la relazione tra episodi di aggiustamento fiscale e il numero di incidenti che indicano instabilità (CHAOS). CHAOS è la somma delle dimostrazioni, disordini, scioperi, omicidi e rivoluzioni avvenute in un solo anno in ogni paese. La prima serie di cinque barre mostra che le frequenze dipendono dalle dimensioni dei tagli di bilancio. Quando la spesa è in aumento, per gli anni di unità di osservazione nei nostri registri dati,  la frequenza dei CHAOS è meno di 1,5 eventi nella media dei paesi osservati. Quando i tagli di spesa raggiungono l'1% o più del PIL, la frequenza cresce a quasi 2 eventi, un relativo aumento di quasi un terzo rispetto ai periodi di espansione di bilancio. Come i tagli si intensificano, la frequenza dei disturbi aumenta. Una volta che le misure di austerità comportano riduzioni di spesa del 5% o più, ci sono più di 3 eventi per anno a paese - il doppio rispetto ai tempi dispesa aumentata.

 

1.    

 

 
Exactly  the  same relationship can be observed  in each of the  four main subcategories of CHAOS. The frequency  of demonstrations, assassinations, and general strikes  rises monotonically with the scale of cuts.  Only in the  case of riots  is there a small decline for the biggest  cut-backs. In the  case of demonstrations, the  frequency  of incidents appears  to rise particularly fast as expenditure cuts  pass the  3% threshold.
 
Earlier  papers  on the  same topic have typically  focussed on case studies,  or on subsets of the developing world. Work  on 23 African  countries  during  the  1980s found that budget  cuts  had typically  no effect on political and  social stability. IMF interventions, on the other  hand,  often led to more frequent disturbances  (Morrison, Lafay,  and  Dessus 1994). Paldam (1993) examines  current account  crises in seven South  American  countries  during  the period  1981-90, using high-frequency  (weekly) data. He finds that the run-up  to new austerity measures  is associated  with  higher levels of unrest,  but  that actual implementationis followed by fewer disturbances. Similarly,  Haggard,  Lafay and  Morrison  (1995) find that IMF interventions and  monetary contractions in developing countries  led to greater  instability. Analysing  the period 1937-1995, Voth  (2011) explores related  issues for the  case of Latin America.  He finds that austerity and unrest  are tightly  linked in a majority of cases. Remarkably, to the best  of our knowledge, there  exists no systematic analysis  of how budget  cuts  affect the  level of social instability and  unrest  in a broad  cross-section of developed countries, over a long period.
 
 
 
Other  related  literature includes work on the  political  economy of fiscal consolidation, and  on its economic effects. The  composition  of fiscal adjustment has been examined; cutting entitlement programs  tends  to produce  persistent improvements in the  budget balance,  while revenue measures and  capital expenditure cuts have only temporary effects (Alesina and  Perotti 1995). The  timing  of stabilization measures  has been explored  in war-of-attrition models, which view relative bargaining strength of different groups as crucial (Alesina  and  Drazen  1991). A rich literature has examined the macroeconomic  effects of budget cuts. Giavazzi  and  Pagano  (1990) and Alesina et al. (2002) find that cuts  can be expansionary. Amongst  the  reasons suggested  for this finding are a reduction in uncertainty about  the  course future spending  (Blanchard 1990a), and  a positive  wealth  shock as a result  of lower taxes  in the future (Bertola and Drazen 1993). Recently,  work by the IMF has suggested  that austerity measures may be less expansionary than  previously  thought; they  may well have the  standard negative Keynesian effects as a result  of lower demand  (IMF  2010; Pescatori, Leigh, and  Guajardo 2011).
We proceed as follows: Section 2 presents  our data,and  section  3 summarizes  our main results.  Robustness checks and  extensions  are discussed in section 4; section 5 concludes.

Esattamente lo stesso rapporto può essere osservato in ciascuna delle quattro sottocategorie principali del CHAOS. La frequenza di manifestazioni, omicidi, e scioperi generali aumenta in modo direttamente proporzionale con la scala dei tagli. Solo nel caso di scontri c'è un lieve calo per lo più nei grandi tagli. Nel caso di manifestazioni, la frequenza degli incidenti sembra crescere in modo particolarmente veloce quando i tagli alla spesa superano la soglia del 3%. 
Precedenti documenti sullo stesso argomento sono tipicamente focalizzati sullo studio di casi, o su sottoinsiemi in via di sviluppo. Il lavoro su 23 paesi africani durante il 1980, ha rilevato che i tagli di bilancio non avevano di solito alcun effetto sulla stabilità politica e sociale. Gli interventi del FMI, invece, hanno spesso portato a disturbi più frequenti (Morrison, Lafay, e Dessus 1994). Paldam(1993) esamina le crisi di conto corrente in sette paesi del Sud America durante il periodo 1981-1990, utilizzando un’alta frequenza  dei dati (settimanale). Egli ritiene che la corsa alle nuove misure di austerità è associata a più alti livelli di agitazione, ma che l'attuazione effettiva è seguita da un minor numero di disturbi. Allo stesso modo, Haggard, Lafay e Morrison (1995) trovano che gli interventi del FMI e le contrazioni monetarie in via di sviluppo hanno portato a una maggiore instabilità. Analizzando il 5°periodo 1937-1995, Voth (2011) esplora le questioni relative al caso dell'America Latina. Egli ritiene che l'austerità e i disordini sono strettamente legate nella maggior parte dei casi. Sorprendentemente, al meglio delle nostre conoscenze, non esiste alcuna analisi sistematica di come i tagli di bilancio incidano sul livello di instabilità sociale e tensioni in un ampio spaccato dei paesi sviluppati, per un lungo periodo. Altra letteratura relativa include il lavoro per l'economia politica di risanamento dei conti pubblici, e sui suoi effetti economici. E’ stata esaminata la composizione dell'aggiustamento fiscale; i  programmi di taglio strutturali tendono a produrre miglioramenti persistenti del saldo di bilancio, mentre le misure concentrate sulle entrate e tagli di spesa in conto capitale hanno solo effetti temporanei (Alesina e Perotti, 1995). La tempistica delle misure di stabilizzazione è stata esplorata con modelli a  “guerra di logoramento”, che constata la relativa forza contrattuale dei diversi gruppi come cruciale (Alesina e Drazen,1991). Una ricca letteratura ha esaminato gli effetti macroeconomici dei tagli di bilancio. Giavazzi e Pagano (1990) e Alesina et al. (2002) trovano che i tagli possono essere decisivi. Tra le ragioni suggerite per questo risultato sono: una riduzione dell’incertezza dispesa sul futuro corso (Blanchard 1990a), e uno shock positivo sulla ricchezza a causa della riduzione delle tasse in futuro (Bertola e Drazen 1993) 3.Recentemente, il lavoro da parte del FMI ha suggerito che le misure di austerità possono essere meno decisive di quanto si pensasse, ma possono anche essere gli effetti negativi degli standard keynesiani a causare la flessione della domanda (IMF 2010, Pescatori, Leigh, e Guajardo 2011). Si procede come segue: la sezione 2 presenta i nostri dati, e la sezione 3 sintetizza i risultati principali. La Verifica della validità e delle estensioni è discussa nella sezione 4, il capitolo 5 conclude. 
 
2. Data
In this  section, we briefly describe our data  and  summarize  its main  features.  We use two datasets – a long-term  one which allows tracing out  the  broad patterns of unrest  and austerity since 1919, as well as a short-term one that contains  richer  information on the causes of unrest.  For  both,  we use information on unrest  as well as on economic performance and  budget measures.

Five main indicators of domestic  conflict in the long-term  data  will form the main  focus of this  study  – general strikes,  riots,  anti-government demonstrations, political assassinations, and  attempted revolutions. These data are part  of the  Cross National Time  Series Dataset,  compiled by Arthur Banks  (2010) and  his collaborators.  The  main  source of data  on unrest  episodes are  the  reports  of the  The New York Times, while the  variables’ definition  is adopted from Rummel  (1974). In addition, we use data  on GDP, government revenue, expenditure, and  the  budget  balance  from a variety of sources. The long-term  data  has information on 26 European countries  and covers the  years from 1919 to 2008. 5

Table 1 gives an overview of the  main variables  and  their descriptive  statistic for the  long-term  data. The average  number  of assassinations and general strikes  was quite  low in our sample,  with  less than  2 events  in each decade. There  were more riots and  more demonstrations – 5-6 per decade. Attempted revolutions are quite  rare,  but  some countries  registered  high levels of instability. The  record in our sample is Germany  in 1923, with  5 recorded attemptsat  overthrow  (with communist insurgencies  in Saxony and Thuringia, the  Hitler  Beer Hall Putsch, and a separatist movement  in the Rhineland). Assassinations and riots similarly  show a broad  range  of observed values.

Using almost  a century  of data  allows us to include some extreme  observations. For example, Austria and Germany  saw major  output declines in 1945 and  1946, respectively.  The biggest reduction in governments spending in our data occurred  in Poland, in 1982; the  second-largest, in Finland,  in 1947. The start of war is often associated  with  big increases  in expenditure. The record-holderin our dataset is Hungary in 1940, with  an increase  of over 30 percent.
 
3  Once the response of labor supply and capital formation is fully taken into account, these effects may not go through (Baxter and King 1993).  
 
 
 
 
4  Data on fiscal variables (Total Central Government Expenditure and Revenue) and GDP are from OECD Stat (2010) for years from 1970 onwards, and from Mitchell (2005) for the period 1919-1970. Data on GDP growth in real terms for the all sample are from Maddison (2010).
5  The 26 European countries included in the long-term data are: Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Netherlands, Norway, Poland, Portugal, Romania, Russian Federation, Serbia, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, United Kingdom, Yugoslavia. 
 
 
 
 
To obtain  a single measure  of instability, we calculate  CHAOS by taking  the sum of the number  of assassinations, demonstrations, riots,  general strikes, and  attempted revolutions. While a crude  way of aggregating indicators, it turns  out  to be powerful. 
In the  robustness section,  we show that alternative methods  of reducing  data  complexity  such as principal  componentsanalysis do not  change our results.
For CHAOS,  the  average  countryin our sample registers  1.5 incidents  per year. Instabilitywas not  constantover time. The  maximum  is higher – Italy  in 1947 saw a total of 38 incidents, including  7 general strikes,  19 riots, and  9 anti-government demonstrations. Figure  1 gives an overview of the evolution  over time,  plotting the  average  of CHAOS as well as the  maximum number  of incidents  observed.  While there  is no clear-cut patternover time, some features emerge. The  interwar period  showed relatively  high levels of unrest,  with  an average  of 2 incidents  per year,  compared  to 1.4 in the  post-war period.  The  immediate post-World War II period,  and  the  period  form 1968 to 1994 also show unusually high levels of unrest.  Comparatively speaking,  the years since 1994 have been unusually  tranquil (average  CHAOS = 0.78)
 
 6  One alternative is the weighted conflict indicator (wci), as compiled by Banks (2010). It gives fixed weights determined to different forms of unrest: Demonstrations have a weight of 200, while political assassinations have a weight of 24.
 
The short-term data  on unrest  is from the  European  Protestand  Coercion Database (EPCD) developed  by Francisco  (2000). The  EPCD  codes daily data  on all reported protest events occurred  in 28 European countries  between  1980 and  1995. The  data  is constructed using the full-text  reports  from more than 400 newspapers  in the  Lexis-Nexis database. We restrict our attention to the same types of protest events  covered in the  long-term  data:  riots, demonstrations, political  assassinations, general strikes,  and  attempted revolutions.7
 
The strength of the  link between  austerity measures  and  unrest  is our first important finding.  Is the  link causal?  Other  factors,  such as generally depressed  economic conditions,  could drive up unrest and  the  need for cut-backs simultaneously. Controlling for economic growth  does not  change  our results.  This  suggests that we capture more than  the  general association  between economic downturn sand  unrest. To demonstrate that causality  runs from cut-backs to unrest,  we refine the data  in two ways: First,  we analyse a more detailed  dataset that gives information about  the  causes of each incident.  Second, we use recently-compiled data  on changes in the  government budget  that follow directly  from policy changes  (Devries  et al. 2011). For  both types  of additional evidence, we find clear indications that the link runs  from budget  cuts  to unrest.  We also conduct placebo  tests  with  other  types  of unrest  – inspired by ecological issues and  world peace, for example  – and  find no effect of budget  measures. Our findings are robust  to a wide range of alternative specifications  and further  tests.  Different  measures  of  unrest  do not  affect our conclusions. We examine if the link between  austerity and  unrest  changes  as countries  institutions improve.  For  most value of the  Polity2  score of institutional quality,  results  are broadly  unchanged.  However, countries with  very high levels of constraints on the  executive  show a weaker degree of association.  Further, we examine  if the  spread  of mass media  changes the  probability of unrest.  This  is not  the  case. If anything, higher  levels of media  availability and  a more developed  telecommunications  infrastructure reduce  the strength of the  mapping  from budget  cuts  to instability. We also test  which part  of the distribution of unrest  is responsible  for our results,  using quantile  regressions: The  higher  the  level of unrest, the  bigger the  relative  impact  of additional budget  cuts.  Finally,  we test  for asymmetries  in the relationship between unrest  and  austerity. Reductions increase  instability, but spending increases do not  cut  the  number  of incidents  to the  same extent.
 
La forza del legame tra le misure di austerità e i disordini è la nostra prima constatazione di irregolarità importanti. È il nesso di causalità? Altri fattori, come generalmente depresse condizioni economiche, potrebbero spingere al rialzo i disordini e la necessità di tagli contemporaneamente. Il controllo della crescita economica non cambia i nostri risultati. Questo suggerisce che è più importante l'associazione generale tra la recessione economica e i disordini. Per dimostrare che relazione di causalità ci sia tra i tagli e i disordini, abbiamo affinato i dati in due modi: in primo luogo, analizziamo un set di dati più dettagliati che fornisce informazioni sulle cause di ciascun incidente. In secondo luogo, utilizziamo i dati recentemente compilati relativi alle variazioni del bilancio statale, che derivano direttamente dai cambiamenti politici (Devries et al. 2011). Per entrambe le tipologie di questi ulteriori elementi di prova, troviamo indicazioni chiare su che tipo di collegamento ci sia tra i tagli di bilancio ei disordini. Stiamo inoltre facendo dei test placebo con altri tipi di disordini - per esempio ispirati a temi ecologici e alla pace nel mondo – senza per altro  trovare nessun effetto collegabile ai provvedimenti di bilancio. I nostri risultati sono validi per una vasta gamma di tecniche alternative e ulteriori test. Diverse misure di disordini non influenzano le nostre conclusioni. Esaminiamo se il collegamento tra i cambiamenti di austerità e di disordini portano ad un miglioramento delle istituzioni dei paesi. Per la maggior parte del valore del punteggio Polity2 di qualità istituzionale, i risultati sono sostanzialmente invariati. Tuttavia, i paesi con livelli molto alti di vincoli dell'esecutivo mostrano un grado più debole di associazione. Inoltre, se prendiamo in esame la diffusione dei mezzi di comunicazione di massa cambia la probabilità di disordini. Questo non è un caso. Se non altro, i più elevati livelli di disponibilità dei media e un'infrastruttura di telecomunicazione più sviluppata riducono la forza della mappatura dai tagli di bilancio all'instabilità. Abbiamo anche prova che parte della distribuzione dei disordini è responsabile per i nostri risultati, utilizzando regressioni quantili: maggiore è il livello di agitazione, più grande è l'impatto relativo dei tagli di bilancio aggiuntivi. Infine, il test per asimmetrie nel rapporto tra disordini e austerità. Le riduzioni della spesa aumentano le instabilità, ma gli aumenti della spesa non tagliano il numero di incidenti nella stessa misura.
 
 
 
The main  advantage of the  EPCD  over the Arthur Banks’ database is that the  former records  the  issue behind  each protest, allowing us to test  the  relationship between  austerity and  unrest  in a very precise way, even if only for a small subset  of the  overall dataset.
There are relatively  few protests that  are caused by austerity measures. At the  same time,  when they  happen,  they  involve a large number  of participants– by far the  largest number  of protesters of any category,  as Table 2 illustrates. These protests tend  to be relatively  peaceful, with  few protesters arrested, injured  or killed, and  relatively  few members  of the security  forces involved.
 
7  We only consider protest events whose number of participants is above 100 for riots and demonstrations and above 1000 for general strikes (no threshold is used for assassinations and attempted revolutions). These are the same threshold used in the Arthur Banks database.


 

In compiling  information on expenditure and  the  budget  balance  data, we need to trade  off the accuracy  of information against  availability over a long time  span.  For  the  1919-2009 dataset, we rely on standard data  sources on the central  government revenue  and  expenditure relative  to GDP  (Mitchell  2007) for the  years 1919 to 1970, augmented by data  from the OECD  (2010) for the period  there after.

Expenditure changes  will serve as the  main  explanatory variable. Figure  2 graphs changes  in expenditure/GDP from one year to the next.  The distribution is almost  symmetric around  the  mean,  with  similar  numbers  of country-years witnessing  expenditure increases and declines (807 vs 685).  In an average  year and  country over the  period,  central  government expenditure relative  to GDP  rose by 0.3%. The  vast  majority of observations falls between increases  and  decreases  of 5%, with  a few outliers  in the  tails  of the distribution (typically driven  by the  beginning  and  end of wars).

 
 
 
In addition, we use the data  by Alesina and Ardagna (2010) for the  cyclically-adjusted budget balance.8  
This has the  advantage of correcting  the  budget  position  for changes  in interest payments and  for the  immediate effect of the economic cycle, which drives both expenditure and revenue without any additional policy decision  being taken. For  a subsample of the  data  (1978-2009, 17 countries), we also use data  by Devries et al. (2011). These authors examine in detail the  policy changes that led to changes  in a country’s  fiscal stance.  Only expenditure cuts or revenue  increases motivated by a decision to press ahead  with  fiscal consolidation are considered. 
Overall,  Devries et al. (2011) find 173 periods of fiscal policy adjustment,
 
As a first pass at  the  data, we repeat  the  exercise in Figure  1 for output growth  (Figure  4). We subdivide the  sample  into terciles,  and  examine how much  the incidence of various  indicators of unrest  declines as growth accelerates.  For  the  summary indicator (CHAOS), there  are a little  more than 2 incidents  when growth  is in the  lowest tercile.  This  falls to 1.3-1.5 incidents  as growth accelerates.  There  is also a clear pattern of decline for demonstrations and  for assassinations. In the  case of riots, the  differences are smaller  overall, where as  in the  case of general  strikes, there  seems tobe little pattern at  all. Based on a first,  visual inspection  of the  data,  it seems that the link between  budget  cuts  and  unrest  is clearer  than  the  one with  growth.
 
8  Alesina and Ardagna use the method of Blanchard (1990b).
9  The approach is similar to the “narrative approach” pioneered by Romer and Romer (1989).
 
 
 
 
 
 
 
Next, we examine  the  correlation  structure of our data  in Table  3. Assassinations, general strikes, riots, revolutions and  demonstrations are all positively and  significantly  correlated with each other.  This  supports our assumption that they  reflect a broader  underlying pattern of social instability and  unrest.  CHAOS is also positively  correlated  with  the weighted  conflict index (wci). Finally,  Table  3 suggests that higher  levels of expenditure and faster  growth  are associated  with  less unrest.  The  simple correlation  of CHAOS with  changes  in the  budget  balance  is positive  and  significant.  Higher taxes  and lower expenditure are associated  with  more unrest,  but  the relationship is not  significant.
 
 
 
 
 
 
 
 
 
In the  case of output changes,  the  coefficient is negative,  but  insignificant  (table  3). The simple correlations suggest  that these  co-movements  do not extend  to all indicators  of unrest  equally – riots,  revolutions, and demonstrations decline as expenditure rises, but assassinations and strikes seem – at  a first pass – uncorrelated. Similarly,  output growth seems to correlate  negatively  with  assassinations, riots, revolutions,  and demonstrations, but  not  with  strikes.  Next,  we examine  the  connection  between budget  position, expenditure, and  unrest  more systematically.
 
 
 
 
3. Results
 
 
The graphical  evidence in Figures  1 and  4 suggests  a link from “hard  times”  – low growth and budget  cut-backs– to unrest.  Next,  we examine  if there  is a systematic relationship between  budget  measures  and  social instability. In this section we also address  the  issue of causality, while in the  next  section  we will test the  robustness of our results.
 
 
A. Baseline Results
We estimate panel regressions of the  type:
 
 
 
 
 
 
 
where Iit  denotes  the  level of instability in country  i at  time t, B is an indicator of the  change  in the  budget  position,  a is a country-specific intercept, l is a time-specific  dummy,  and  X’ is a vector  of control  variables.
 
We use CHAOS as the  dependent variable  in our baseline  specification, and  test  the robustness of findings to alternative specifications later.  Table  4 gives the  main  results.  Under OLS with  fixed effects and year-dummies, we find that expenditure increases  reduce instability in a powerful way (column 1). A one standard-deviation increase in expenditure cuts  the  number  of incidents  (CHAOS) by 0.4 per year and  country. Tax  increases  have a positive sign, but  the  effect is not significant at  standard levels of rejection  (column  2).
It is also small – a one standard deviation  rise in the  tax/GDPratio  increases unrest  by less than  0.01 events.  Overall,  we find that improvements in the budget  balance  raise the  level of unrest  (column  3). As the  results  in columns (1) and  (2) make clear, this  reflects the impact  of expenditure cuts,  and  not  of tax  increases. CHAOS is a count  variable.  Hence, the  use of OLS may not  be appropriate.Columns  (4)-(6)  give the results  for Poisson  Quasi-Maximum Likelihood estimation, with  fixed effects. We find the  same patternas before, with  strong  effects for expenditure cuts and  much  weaker ones for tax increases.10
 
10  We also experimented with using negative binomial regressions, but results were largely unchanged.
 
 
Which component of CHAOS is responsible  for the  significant  predictive power of budget  cuts?  In Table  5, we use the  same specification  as in Table  4 under  Poisson QML, looking at  the  effect of expenditure cuts  on each of the components of the aggregate indicator  of instability – general  strikes, demonstrations,riots, assassinations, and  attempted revolutions. Out  of the five outcome  variables,  four show the expected sign, and  all of them  are statistically significant.  The  only variable  that does not  show a large, significant coefficient is general strikes. On average,  years with  expenditure increases  showed fewer general strikes, but  there are numerous  general strikes thatare not  an immediate reaction  to economic conditions  and  budget  measures (such as, for example, the 1926 general strike  in Britain). For the other  variables,  the coefficients are large, indicating  that austerity measures coincide with  significant increases in demonstrations, attempted revolutions, riots, and assassinations. In all specifications, the effect of GDP growth  on unrest  is negative.  In contrast to the results for expenditure changes, the  effect is not  tightly  estimated, except  in the case of demonstrations, when it is also large – every 1% increase  in GDP  cuts  the  number  of demonstrations by close to 0.4 events.
 
 
 
Table 6 takes this analysis one step further, by breaking  the  period  1919-2009 into  four sub-periods.  We distinguish the  interwar period  from the  period  of immediate  post-World War II reconstruction, the  period of slowing growth into  the  1980s, as well as the years after the  fall of the Berlin Wall after 1989. On the whole, we find the same patternas in the sample as a whole, with  the exception  of the  last  two decades. The effect of changes  in budget expenditure on unrest  is strongest in the  tumultuous interwar years,  when the  estimated coefficient is fifty percent  larger  thanin the  sample  as a whole. The effect of GDP  growth  is negative,  but  not  tightly  estimated.In the  years after  1945, the inverse relationship between  expenditure and  unrest  remains.  Strikingly,  however, more growth  now appears  to lead to more unrest.  While it is difficult to test  for the causes of this reversal  exactly,  it seems thathigh rates  of output growth  may have encouraged worker militancy more generally.  At a time  when many countries  reached  full employment, this  effect seems to have become dominant. The normal  patternof GDP  growth  reducing  unrest  reasserts  itself after  1965, when there  is also still a clear negative  effect of higher  government expenditure.
 
The fall of the  Berlin wall saw the spread  of Western-style democracy eastwards. The overall connection  between  austerity and social instability now changes  sign, and  becomes in insignificant.  This  suggests  to us that non- economic causes became a dominant feature  of the period.  Below, we examine the  issue in more detail  with  the  help of a dataset that allows us to look at the motive of each demonstration.
 
 
 
B. Causality
The obvious challenge  in interpreting (1) is the potential for omitted  variable  problems.  Itis possible thatthe  economic cycle is simultaneously driving  both unrest  and  the  need for budget cuts.  Above, we already  control  for GDP growth  rates,  and  our main  finding remains unaffected.However, theomitted  variable  problem  would only be solved if we measured  the effect of economic outputon instability perfectly.  Since this is unlikely, we present  a different add  two type  of analysis.  We use a related  dataset that offers detailed  information,for a shorter  time  period,  on the  causes  behind  each  unrest  event.  This  allows us todemonstrate the  connection  between  social instability and  expenditurecuts  more directly.
As described  in the  data  section,  the  EPCD’s  datasetallows us to pin down the mainmotive  behind  each public demonstration.We examine  if the public  assemblies that are motivated by anti-austerity sentiment – as determinedby the  newspaper  records  in Lexis-Nexis – are significantly affected by actual  changes  in fiscal policy. Our  approach  here is similar  to what  has been called the “narrative approach”  (C.D.  Romer  and  D.H. Romer 1989). Table  7 gives the results.  If we use the  same specification  as in Table  1 (where we analysed  the datasetspanning  the  period  1919-1999), we find similar  results. Increasing  expenditurelowers levels of unrest  (column  1). The key variable  driving  the relationshipbetween  budget  balance  and  instabilityis expenditure,not  taxes  (columns  2 and 3). The  results  are robust  to including countryand  year fixed effects. In column  6, we investigatewhat  happens when we use all forms of demonstrations, not  just  those  associated  with austerity. Thecoefficient is small, positive,  and  insignificant.

 

We can strengthenthis  result  further  by conductinga placebo  test.  In Table 8, we use a set of alternative types  of unrest,  and  test  if they  can be predictedby the  same explanatory variables  as in Table  7. Labour  disputes  and  unrest  inspired  by the  state  of the economy are more frequent  when budgets  are being cut,  but  the  link is not  strong  or statisticallysignificant.  Peace rallies, and unrest  as a result  of educationissues, show the opposite  sign of the coefficient on austerity – times  of rising expenditurealso seem to bring  these issues to the fore. Overall,  the  placebo test  shows thatonly in the  case of anti- austerity demonstrations is there  a strong  and  significant link with changes  in government expenditure.

Another  way to strengthenthe  argumentfor a causal  link is to examine budget  measures  in more detail.  Some of the  variationin the  budget  balance thatwe have used so far will simply reflect revenue  and  expenditure changes thatare driven  by the  economic cycle. A simple way to deal with  the  problem is to use Alesina and  Ardagna’s  (2010) cyclically-adjusted primary  budget  balance.  In table  9, col. (2), we report  the  results.  The  coefficient on budget  changes  is almost  identical  to the baseline specification.  In col. (3), we use the IMF measure  of policy-action  based changes  in the budget  balance.11  

This also produces  a large, significant  coefficient. The  closer we get to measuring  the impact  of policy measures,  the  larger  coefficient becomes. This  strengthens the case for a causal  link between  unrest  and  austerity.

11  Since Devries et al. (2011) only report positive changes in the budget balance, data from IMF International Financial Statistics has been used to proxy for negative changes in the budget position in the IMF (2011) series, sign and size of the coefficient are not affected by this assumption.
 
 
 
 
4. Robustness and  Extensions
 
In this  section, we examine  the  sensitivity of our results.  We first examine interaction effects with institutional factors.  Do countries  with  more accountable governments weather  the storms  of austerity better?. We also examine  if the  effect may be driven  by outliers,  whether positive  or negative changes  in expenditure matter more for the  effect on unrest,  and whether the effect is constantin all parts  of the distributionof the  dependent variable.
Greater constraints on the  executive  and  more democracy  should  on the  hand  - reduce  social conflict; on the  other,  there  will be less repression  by the  authoritiesas Polity scores improve.  Which  effect dominatesis not  clear ex ante.  Table 10 demonstrates that in countries  with  better institutions, the responsiveness  of unrest  to budget  cuts is generally lower. Where  constraintson the  executive  are minimal,  the coefficient on expenditure changes is strongly  negative  – more spending  buys a lot of social peace. In countries  with Polity-2 scores above zero, the  coefficient is about  half in size, and  less significant.  As we limit  the sample  to ever more democraticcountries,  the  size of the  coefficient declines. For  full democracies  with  a complete  range  of civil rights,  the  coefficient is still negative,  but  no longer significant.
The link with  growth  is less clear-cut.  Higher outputhardly  dents  the tendency  to riot,  demonstrate,assassinate,  or strike  in countries  with  low institutional quality.  The  opposite  is true  on average  in countries  with  scores above zero, and  throughoutthe  range  of scores. The  only exception  is for full democracies, where the connection  is weaker.

 

 

When does the  link between  budget  cuts  and  unrest  become particularly strong?  We examine which part  of the  distribution of CHAOS shows a particularly large impact  of austerity measures.  To do so, we estimate  quantile  regressions,  where we estimate  the conditional median, and  then  the  effect from the  5th  to the  95th  percentile  of the  distribution of CHAOS.  Figure  5 shows the  size of effects. The  estimated coefficient is zero for much  of the range.  Only from the  80th  percentile  upwards  – for country-year observations with  two or more incidents  – is the effect visible. It then  grows rapidly  as estimated coefficient on expenditure changes (and on outputgrowth)  increases at higher  and  higher  percentiles  of the distribution  of CHAOS.  This  suggests that unrest  reacts  particularly strongly  to budget  cuts  and  growth  when unrest  levels are already  high.
 
 
How much  does our main  finding depend  on the  way in which we aggregate  unrest? CHAOS is the  simple sum of incidents. Instead, we can use the weighted  conflict index, as compiled by Banks (1994) and  collaborators.It encompasses  a larger  set of domestic conflicts including,  in additionto the componentsof CHAOS,  purges,  major  government crisis and  guerrilla  warfare. It also assigns different,  fixed weights  to each individual  component. The correlation  coefficient of the  variable  with  CHAOS  is 0.75, significant  at  the 1% level. Anotheralternative is touse the  first principal  component of the five indicators that go into  CHAOS. They all enter  with a positive  weighting.  The  first principal  component explains  0.42 of the  overall variance.  The correlationcoefficient with  CHAOS is 0.98.
In Table  11, we use both  wci and  the  first  principal  as dependentvariables.  Since the dependentvariable  is no longer a count  variable,  we use panel  OLS, and obtain  large and significant  coefficients for expenditurechanges  and  the budget  position.  As before, the  same is not  true  for tax  changes.  The  results are largely identical  in terms  of magnitude and significancewith  the  baseline results  in Table  3. We conclude that  the way in which we measureunrest  does not  matterfor our main  finding.
 
 
 
 
 
An additionalfactor  thatcan be questioned  involves the  use of the  sum of unrest  in the baseline  results.  The variable  CHAOS is designed to capturethe intensity of unrest,  but it may be thatit is influenced  by a number  of outliers with  a high count  of  incidents.This would thenmake it easier to find significant  effects. To examine  this potential issue, we transformCHAOS into a simple dichotomousvariable,  with  unrestcoded as equal to unity  if there  are one or more incidents  in a country  in a single year.  In table  12, we re-estimatethe  baseline  regression  with panel logit using country-and  year-fixed  effects.
 
 
We find the  same results  as before – expenditurecuts  wreak havoc,  tax increases  do so only toa small extent  and  insignificantly. Overall,  the  budget  balance  mattersfor predicting unrest.  We conclude  thatthe  role of outliers  is not  decisive in underpinningthe  relationship we establishedin baseline results.
 
 
 
 
 
Which part  of the  variationin the  explanatory variables  is responsible  for the link between austerityand  unrest?  Do increases in expendituredo as much  to reduce  unrest  as cuts increasethem?  In Table  13, we look at  the  issue. Column  (1) shows the  results  for expenditure changes  thatare positive.  The coefficient is negative,  but  not  large, and  not significant.  In contrast,if expenditurechanges  are negative,  they  matter a great  deal for unrest,  driving up CHAOS by 0.19 incidents  for each standard deviation  of expenditure cuts. Next,  we repeat  the  exercise for outputchanges.  Increases  in output do much to cut  unrest (col. 3), with  a one standard deviation  increase in output (3.77%) reducing  CHAOS by 0.2 incidents  on average.  In contrast,declines do not  set off major  disruptionsto the  same degree. Overall,  the  results  in table 12 confirm thatthe  relevant  identifying  variation for expenditure changes comes from cuts;  for outputchanges,  it comes from positive  growth,  not recessions.
 
 
 
 
 
 
Does greater  media  penetrationincrease  or reduce  unrest?Events  in the  Arab world in 2010 and  early 2011 have led many to believe thatgreater  media availability  tightensthe link between discontentand  unrest.  Data  on media
 
 
penetrationis available  in the  Banks  dataset. Four  indicatorsare suitable  – phone penetrationsper capita,  radio  and  television  take-up,and  the  number  of telegrams  sent  per capita.  Radio  and  television  are unidirectional forms of media,  allowing typically government-controlled messages to be broadcastto the  population.If anything, they  should make it easier for authoritiesto reduce  unrest.  Phones  and  telegrams,  on the  other  hand, allow peer-to-peer communication. All else equal, the  expected  effect is thatthey  facilitate organized  protest.
To analyse  the  data, and  to avoid confusing results  with  the  growing availability of broadcasting and  telecommunications over time,  we rank penetration rate  in our sample ineach year.  We do separately for each category,  and  then  sum the  ranks  for each country-year.This  gives a rank ordering  of media  penetration in year y. We thendivide the sample  at  the median.  Table  14, col. (1) and  (2) presents  theresults.  We find that below- average  media  penetrationis associated  with  a strong  effect of expenditurecuts  on unrest.  Above the  median,  the  effect disappears.There  is also some evidence that  the opposite  pattern obtains  with  respect  to economic conditions  – the  responsiveness  to output changes  increases  as media penetration grows.In col. (3)-(6),  we differentiatebetween  uni-directionalinformation media (infomedia)  and  peer-to-peer  telecommunications (peermedia). While there is some attenuation of the  effect of expenditure changes, it is milder  than  for all media.  For  both  types,  the  effect of economic conditions changes  from insignificant  (in the  part  of the  sample  with below- median  penetration) to highly significant (above the  median).  These  results  do not  suggest thatcountries  which, at any one point  of time,  have greater  availability  of mass media  (relative  to their  neighbors) experience  a higher level of unrest.12
 
 
5. Conclusions
The political  economy literature on austerity suggests a paradox.There  is no significant punishmentat  the  polls for governments pursuing  cut-backs(Alesina,  Perotti,and  Tavares 1998; Alesina,  Carloni,  and  Lecce 2010), and  no evidence of gains in response  to budget expansion  (Brenderand  A. Drazen 2008).  Also, the  empirical  evidence on the economic effects of budget  cuts  is mixed,  with some studies  finding an expansionaryeffect, and  others,  a contractionaryone.13  Why, then, is fiscal consolidation  often delayed,  or only implemented half-heartedly?
This paper  suggests  one possible reason  why austerity measures  are often avoided – fear of instability and  unrest.14  
Expenditurecuts  carry  a significant  risk of increasing  the  frequency  of riots,  anti-government demonstrations,general strikes,  political  assassinations,and  attempts at revolutionary overthrow  of the  established order.  While these  are low- probabilityevents  in normal  years, they  become much more common as austeritymeasures  are implemented. This  may act  as a potentbrake  on governments.In line with  our results  on expenditure,Woo (2003) showed that countries  with  higher  levels of unrest are more indebted.High levels of instabilityshow a particularlyclear connection  with  fiscal consolidation.
 
We demonstrate that the  general pattern of association  between  unrest  and budget cuts  holds in Europe  for the  period  1919-2009. It can be found in almost  all sub-periods,  and for all types of unrest.  Strikingly,  where we can trace  the  cause of each incident  (during  the period  1980-95), we can show thatonly austerity-inspired demonstrations respond  to budget cuts  in the  time- series. Also, when we use recently-developeddata  thatallows clean identificationof policy-driven  changes  in the  budget  balance,  our results  hold. Finally,  the results  are not  affected by using alternativemeasures  of unrest.  Contraryto what  might  be expected,  we also find no evidence thatthe  spread of mass media facilitates  the rise of mass protests.

 
12  The obvious alternative is to condition on the absolute level of, say, phone penetration. Most of the variation  in  phone  penetration,  however,  simply  reflects  GDP  growth  and  the  declining  cost  of telephones relative to all other goods; no clear pattern emerges.
13  Alesina and Silvio Ardagna 2010; Alesina, Silvio Ardagna, et al. 2002; Pescatori, Leigh, and
Guajardo 2011. An early example in the literature is Giavazzi and Pagano (1990).
14  Alesina, Carloni and Lecce (2010) also suggest that implementation of budget measures may be harder if the burden falls disproportionately on some groups. War-of-attrition models of consolidation
are one alternative (Alesina and Drazen 1991).

 
 
References:
 
Acemoglu,  Daron,  and  James  Robinson.  2000. “Why  did the  West  Extendthe Franchise? Democracy,  Inequality, and  Growth  in Historical Perspective.” Quarterly Journal  of Economics 115 (4): 1167-1199.
Alesina, Alberto,  Silvio Ardagna,Roberto  Perotti,and  F. Schiantarelli.  2002.
“Fiscal policy, profits,  and  investment.” American Economic Review 92 (3): 571–589.
Alesina, Alberto,  Dorian  Carloni,  and  Giampaolo  Lecce. 2010. “The  electoral consequences of large fiscal adjustments.”Harvard  University, mimeo.
Alesina, Alberto,  Roberto  Perotti,and  Jose Tavares.1998. “The  political economy of fiscal adjustments.” Brookings Papers on Economic Activity 1998 (1): 197-266.
Alesina, Alberto,  and  Silvio Ardagna.2010. “Large changes  in fiscal policy:
taxes versus  spending.”  Tax Policy and The Economy.
Alesina, Alberto,  and  Allan Drazen.  1991. “Why  are Stabilizations Delayed?”
The American Economic Review 81 (5) (December):1170-1188.
Alesina, Alberto,  and  Roberto  Perotti.1995. “The  political  economy of budget  deficits.” Staff Papers-InternationalMonetary Fund 42 (1): 1-31.
Banks, ArthurS. 1994. Cross-National Time-Series  Data  Archive.
Binghamton,New York.
Baxter,M., and  R.G.  King. 1993. “Fiscal policy in general equilibrium.”  The
American Economic Review: 315-334.
Bertola,  Giuseppe,  and Allan Drazen.  1993. “Trigger  Points  and  Budget  Cuts: Explainingthe Effects of Fiscal Austerity.” American Economic Review
83 (1): 11-26.
Blanchard,Olivier.  1990a. “Commenton Giavazzi  and  Pagano,‘Can  Severe
Fiscal ContractionsBe Expansionary?’” NBER macroeconomics annual
5: 111-116.
———. 1990b. “Suggestions  for a New Set of Fiscal Indicators.” OECD Economics DepartmentWorking Papers.
Boix, Carles.  2003. Democracy  and redistribution. Cambridge, UK. Cobban, A. 1964. The social interpretation of the French Revolution.
Cambridge university press.
Devries,  P.,  J. Guajardo,D. Leigh, and  A. Pescatori.2011. “A New Action- based Dataset of Fiscal Consolidation.” IMF Working Paper  No.
11/128.
Doyle, W. 2001. The French Revolution: a very short introduction.  Vol. 54.
Oxford University  Press,  USA.
Giavazzi,  Francesco,and  Marco Pagano.1990. “Can  Severe Fiscal Contractions Be Expansionary?Tales  of Two Small EuropeanCountries.” NBER Macroeconomics Annual: 75-111.
Haggard,  S., JD  Lafay,  and  C. Morrisson.  1995. The political feasibility of
adjustmentin developing countries. OECD.
Hunt,  L.A. 2004. Politics, culture, and class in the French Revolution. Vol. 1.
Univ of California  Pr.
IMF. 2010. Will It Hurt?  Macroeconomic  Effects of Fiscal Consolidation.In .
World Economic Outlook.  IMF.
Mitchell,  B. R. 2007. InternationalHistorical Statistics. 6th  ed. Basingstoke,
Hampshire:  Palgrave Macmillan.
Morrison, C, JD  Lafay,  and  S Dessus. 1994. “Adjustmentprogrammes  and politico-economic interactionsin developing  countries:  lessons from an empirical  analysis  of Africa in the 1980s.” From Adjustment  to Development in Africa: Conflict, Controversy, Convergence, Consensus.
OECD.  2010. OECD Economic Outlook No. 89. Paris.
Pescatori,A., D. Leigh, and  J. Guajardo.2011. “Expansionary Austerity New
InternationalEvidence.”
Romer, C.D.,  and  D.H. Romer.  1989. Does monetarypolicy matter?A new test  in the spiritof Friedman  and  Schwartz.  In NBER Macro Annual.
MIT Press.
Rummel,  R.J.,  R. Tanter,and  Inter-university Consortium for Political  and Social Research.  1974. Dimensions of conflict behavior within and between nations, 1955-1960. Inter-University Consortium for Political
Research.
Soboul, A. 1974. The French Revolution, 1787-1799: from the storming of the
Bastille to Napoleon. Random  House.
Voth, Hans-Joachim. 2011. TighteningTensions:  Fiscal Policy  and  Civil
Unrest  in Eleven South  American  Countries,1937-1995. In Fiscal
Policy and Macroeconomic Performance. Central Bank  of Chile.
Woo, J. 2003. “Economic,  political,  and  institutional determinants of public deficits.” Journal of Public Economics 87 (3-4): 387-426.

 

 

 

 

Licenza Creative Commons

Permissions beyond the scope of this license may be available at f.marinelli@alice.it

 

Sfoglia luglio        settembre
Capitoli:
Consiglio i siti:
Biografia e ideali:
Laureato a Bologna. Mi affascina occuparmi come hobby della sociologia e filosofia, tenendo ben presenti i problemi della natura, che soffre oltremodo per la presenza asfissiante dell'uomo. Il mio ideale è trovare una forma di convivenza degli uomini che: 1) abolisca le guerre; 2) promuova uno sviluppo sostenibile; 3) trovi un equilibrio permanente con la natura del pianeta Terra; 4) ridistribuisca le risorse tra tutti gli abitanti del pianeta; 5) aumenti le risorse relative su scala mondiale, mediante diminuzione della popolazione con un rientro morbido sotto i 4 miliardi, prima della fine del petrolio. "Imagine there's no countries It isn't hard to do, Nothing to kill or die for And no religion too. Imagine all the people Living life in peace... You may say I'm a dreamer But I'm not the only one. I hope someday you'll join us And the world will be as one. Imagine no possessions, I wonder if you can, No need for greed or hunger A brotherhood of man. Imagine all the people Sharing all the world..." Imagine di John Lennon
Sondaggio:
Per la SOPRAVVIVENZA del BLOG: CLIKKAMI per fare una DONAZIONE a piacere mediante l'uso di PayPal
Proponi su Blog News
Donazione con PayPal
Guest Book
Ricerca Google in Sito
------------------------------------
--- Primi ingressi nel blog ---
------------------------------------
Ingressi TOTALI: 1326590

Per MIGLIORARE il BLOG: CLIKKAMI per scrivere sul GUESTBOOK
------------------------------------
Support Wikipedia
------------------------------------
Clikkami per ricerca Google dentro il sito
------------------------------------
adv