Prof Vijay Joshi - LK Jha Memorial Lecture at Reserve Bank of India

Vijay Joshi @MertonCollege @UniofOxford & author of '#India's #LongRoad' https://t.co/T9wvw2MctQ … … also gave the @RBI #LKJha #MemorialLecture - a rare #GlassHalfEmpty #OneHandedEconomist's analysis of India's policy challenges: the main one, #EmploymentGeneration, requires:

— Satyen Baindur (@Satyen_Baindur) April 30, 2018

...increasing the #LaborIntensity of India's Growth, and esp its #Exports - which requires both #LaborMarketReform (largely outside @RBI’s remit) +#ExchangeRate management – diagnoses the 2004-17 exchange rate ↑ as having been caused by the #DutchDisease – large capital inflows; pic.twitter.com/shMDXD1iFM

— Satyen Baindur (@Satyen_Baindur) April 30, 2018

...and suggests ExchRate was allowed to remain high mainly to keep domestic inflation in check... Recommends departing 4m strict #InflationTargeting & separating monetary & exchange rate policy using #SterilizedIntervention & #TargetedCapitalControls 2 prevent exchange rate ↑.

— Satyen Baindur (@Satyen_Baindur) April 30, 2018

Will Eurozone Quantitative Easing End in Time for Monetary Policy Normalization?

#QuantitativeEasing in the #Eurozone may continue till 1Q2019, which means the first rate increases, which can only come after it ends, may leave rates too low to have an impact if @ecb needs to lower them further @ the next downturn. via @elerianm @FT https://t.co/YwBDDCFmY3

— Satyen Baindur (@Satyen_Baindur) April 26, 2018

Artificial Intelligence, Machine Learning, Cybernetics and all that...

Prof Michael Jordan @UCBerkeley #CS #Statistics #MustRead on AI 1)Not '#AI' but #IA Intelligence Augmentation & #II Intelligent Infrastructure 2)Challenges of #IA & #II r not subsumed by human-imitative #AI 3) #MachineLearning is a branch of statistics... https://t.co/AiwMj8unlH

— Satyen Baindur (@Satyen_Baindur) April 24, 2018

4) John McCarthy coined '#AI' to distinguish his approach 4m Norbert Wiener's 'cybernetics'; 5)Ironically, what people call '#AI' now is traceable to Wiener's ideas 6)#backprop was *rediscovered* in the '80s, a control-theoretic version was in use much earlier, eg w @NASA Apollo

— Satyen Baindur (@Satyen_Baindur) April 24, 2018

ArchaeoGenetics of SouthAsian and European Populations

#PaleoAnthropology #ArchaeoGenetics #EthnoLinguistics…. #GeoPolitics(!?) @lizzie_wade @sciencemag @biorxiv https://t.co/ruu7aejhvk https://t.co/rQdSrQ96Jo on ‘A Tale of Two #SubContinents’ - #India & #Europe…the peopling of the two occurred via parallel (if not exactly coeval) pic.twitter.com/ZOFwgrilut

— Satyen Baindur (@Satyen_Baindur) April 23, 2018

...processes – in which steppe herders (4m present-day Kazakhstan, home 2 the original #IndoEuropean speakers) mixed w farmers 4m #Iran & then moved both Westward 2 #Europe & Southeastward 2 India to further mix w populations already there - extent of admixture perhaps varying...

— Satyen Baindur (@Satyen_Baindur) April 23, 2018

GeoPolitics and the India-Iran Chabahar Port Project

#GeoPolitics In the @CSIS report ‘China’s Maritime #SilkRoad’ (#CMSR) Prof Harsh Pant @KingsCollegeLon discusses the #India-#Iran #Chabahar Port Project (which potentially yields an #India-#Europe land route that bypasses the #SuezCanal)… pic.twitter.com/AxOIYOVoe3

— Satyen Baindur (@Satyen_Baindur) April 21, 2018

...and how #China could end up being in it – in part because Iran’s vision (in which #Teheran is @ the center of a Vienna-Beijing rail line) blends better w China’s overall vision under #CMSR & #OBOR, but also because of #China's gr8r experience in delivering on such projects... pic.twitter.com/wSMpwi424S

— Satyen Baindur (@Satyen_Baindur) April 21, 2018

via @gurmeetkanwal @seemasirohi #GreatMapshttps://t.co/HIYc3vfG6s pic.twitter.com/EK55J7LH9i

— Satyen Baindur (@Satyen_Baindur) April 21, 2018

The Lords' AI Committee Report - 16 April 2018

.@LordsAICom foresee the possibility of inadvertent or intentional harm resulting from #AI or #AI-powered applications, but disfavor #AI-specific #regulation, in part because it could be counterproductive, for example, firms might attempt to avoid it by not labeling products 'AI' pic.twitter.com/deUVxgYlAY

— Satyen Baindur (@Satyen_Baindur) April 16, 2018

… While some witnesses argued 4 a #TechnologyNeutral approach, implying the sufficiency of existing regulation, @LordsAICom recommended that the Centre for Data Ethics & Innovation; & the UK Government Office of AI - identify gaps in existing regulation, if any. #LordsAIReport pic.twitter.com/kizeFErz5j

— Satyen Baindur (@Satyen_Baindur) April 16, 2018

The @LordsAICom recommend that a Global Summit on #AI convene in London in 2019, to develop a common ethical framework for development/deployment of #AI. And interestingly, 11/13 Members of @LordsAICom made 'Declarations of Interest' in orgs & companies working with and on #AI!

— Satyen Baindur (@Satyen_Baindur) April 16, 2018

Bank of England's Andy Haldane on Cambrian Explosion in Macroeconomic Modeling

.@BoE_Research Andy Haldane makes case for #CambrianExplosion in #Macroeconomic models – for Dynamic Stochastic General Equilibrium (DSGE) models, the critique was not that they did not forecast #GFC; they could not conceive of endogenous #FinancialCrisis at *any* time! Figure below shows GDP forecasts from 27 forecasters incl #BoE 2007Q4 w actual data at end 2015 pic.twitter.com/g3jk3B1tAl

— Satyen Baindur (@Satyen_Baindur) April 13, 2018

Might need separate models for i) near-equilibrium fluctuations w approximately linear dynamics ii) far-from-equilibrium (or multiple equilibria) with order parameters like unemployment/GDP undergoing discontinuous phase transitions & heterogeneous agents to explain endogenous crises like the Global Financial Crisis pic.twitter.com/VFZqnA7DRf

— Satyen Baindur (@Satyen_Baindur) April 13, 2018

Climate Change is a Central Bank Issue

.@FerdiGiugliano Global Warming Is a #CentralBank Issue: Assessing #ClimateRisks, eg those on #Insurers' books- is w/in the #FinancialStability mandate: may require extending the #SupervisoryHorizon and possibly #Compulsory #Transparency https://t.co/YB9jvyICmF via @HalfonJulien

— Satyen Baindur (@Satyen_Baindur) April 13, 2018

#Villeroy - And on top of that, supervisers « should develop forward-looking carbon stress tests for both insurance companies and banks – what I call “the video of risks”» 2/2 https://t.co/p6p0AtEKiF

— Banque de France (@banquedefrance) April 6, 2018

New IMF Global Macrofinancial Model - II

New #IMF Global #Macrofinancial Model #GFM: #Impulse #Response How does #ConsumptionrPriceInflation respond to a 1% rise in the US monetary policy rate? In #SaudiArabia (blue); #SouthAfrica (black) #Poland (red) and US itself (grey); Y-axis in percent; X-axis - time, in quarters pic.twitter.com/fDNfm6vMLA

— Satyen Baindur (@Satyen_Baindur) April 11, 2018

...and looking at effect of a shock in a #MacroPru type of instrument: How does Output (#GDP) change with a 1% reduction in the #Domestic mortgage #LoanToValue #LTV ratio? In #China (blue); In #Spain (black); in #Ireland (red) and in the US itself (grey).. pic.twitter.com/ix2BSEsqiw

— Satyen Baindur (@Satyen_Baindur) April 11, 2018

New IMF Global Macrofinancial Model

#FrancisVitek #IMF #Monetary & #Financial #Contagion in the new #IMF #Global #MacroFinancial Model: #AEs modeled as #InflationTargeting Economies; #India & #China as Managed #ExchangeRate regimes. #CapitalMarket & #Interbank #Contagion channels r modeled separately; Via @MacroPru

— Satyen Baindur (@Satyen_Baindur) April 11, 2018

#China taken as Low #Contagion economy both channels; #India taken as High #CapitalMarket #Contagion. #IMF new #Global #Macrofinancial Model, (abbreviated #GFM); parameterized both by #Calibration & #Estimation, #DynamicStochasticGeneralEquilibrium #DSGE; 40 national economies..

— Satyen Baindur (@Satyen_Baindur) April 11, 2018

IMF WEO April 2018: Technology Diffusion and Productivity Growth

#IMF #WEO April 2018 Chapter 4: Annual #Productivity Growth in #EmergingMarketEconomies #EMEs over 1995-2014 is ~2%; and for #AEs ~1.5%. But even 4 #EMEs, #Domestic R&D makes a *larger* contribution to #Productivity Growth than #Foreign R&D. R results robust 2 excluding #CHN ? pic.twitter.com/brgfS74LzP

— Satyen Baindur (@Satyen_Baindur) April 10, 2018

Using an augmented #EME data set, authors find Annual #Productivity growth during 1995-2003 to be ~4%, but declining to less than 2% during 2004-14; however, #Foreign R&D contributed a larger share, both relative to 1995-2003, and the smaller data set of the previous figure… pic.twitter.com/L2JTI9ftZH

— Satyen Baindur (@Satyen_Baindur) April 10, 2018

IMF World Economic Outlook April 2018 - Chapter 2

.@LounganiPrakash “Recessions are not rare,” echoed Prakash Loungani,  a macro-economist at the #IMF. “What is rare is a recession that is forecast in advance.” Despite an increased amount of economic data..., “the ability to predict downturns remains dismal”, he told @FT

— Satyen Baindur (@Satyen_Baindur) April 9, 2018

Inability to explain the #LFPR decline 4 most countries by 'exogenous' effects #Demographics #AgeingPopulation or #Cyclical Effects (#GreatRecession, modest recovery) => role 4 policy interventions – e.g. unemployment benefits, taxes, coordination of wage setting, migration.

— Satyen Baindur (@Satyen_Baindur) April 10, 2018

Difficulty of Economic Growth Forecasting and Data Revisions

.@BEA_News US GDP Growth data Q12015 illustrate both difficulty of forecasting & the multiple revisions data undergo...#IMF April #WEO better than Oct... Economic forecasts shd provide “broad trends, not specific numbers” Mr Donovan @UBS. ht @Went1955 @FT https://t.co/rVXNVVQUQI

— Satyen Baindur (@Satyen_Baindur) April 9, 2018

.@LounganiPrakash “Recessions are not rare,” echoed Prakash Loungani, a macro-economist at the #IMF. “What is rare is a recession that is forecast in advance.” Despite an increased amount of economic data..., “the ability to predict downturns remains dismal”, he told @FT

— Satyen Baindur (@Satyen_Baindur) April 9, 2018

Inter-Country Economic Growth Correlations and Globalization

...for #AEs, the #DebtLimit increases till all debt is #GDP-linked, but 4 #EMs, the marginal effect on #DebtLimt begins to decline after ~60% debt is #GDP-linked...some caveats apply: #MoralHazard #RiskAppetite etc & the analysis may need 2 b modified 4 cntrys in a #CurrencyUnion

— Satyen Baindur (@Satyen_Baindur) March 29, 2018

.@jeanboivin @IMateosylago @blackrock look at #GDP #Growth 1978-2017 4 the 18 largest #AEs & #EMs, find – ~half of individual country variance in #GDP Growth rates (18 largest #AEs & #EMs) can b explained by a single common factor that is highly correlated w ‘business confidence’ pic.twitter.com/JVYR1T1nFV

— Satyen Baindur (@Satyen_Baindur) April 2, 2018

Growth 1998-2017 is more highly correlated than 1978-97. Impact of #GeoEconomic events eg #EUExpansion #ChinaAccession 2 #WTO, '97 #AsianCrisis, #NAFTA clearly seen. #Protectionism can harm future #Growth not just via the trade channel, but also through the #Confidence channel.

— Satyen Baindur (@Satyen_Baindur) April 2, 2018