20 febbraio 2012

Turbofinanza?

Il grafico in basso (click per ingrandire) mostra l'andamento tra il 1975 e il 2010 di una misura media della correlazione tra gli indici azionari di sei importanti mercati (valori mensili, USA, Giappone, UK, Germania, Francia, Canada). Le aree in grigio nella figura denotano le fasi di recessione negli USA così come identificate dal National Bureau of Economic Research.

Emergono diversi spunti interessanti. Innanzitutto, le fasi di rallentamento economico più pronunciato e quelle di maggiore turbolenza finanziaria sono caratterizzate da un sistematico aumento della correlazione. Cioè i mercati tendono a muoversi maggiormente all'unisono in queste fasi. Poi, si nota un robusto aumento della correlazione negli ultimi 15 anni.

Per chiarire le cause sottostanti entrambi i fenomeni, assolutamente cruciali, è necessario fare un'analisi quantitativa molto più approfondita di quanto si possa condurre qui (me ne sto occupando con Anton Muscatelli della University of Glasgow). Essenzialmente, occorre capire la misura in cui le variazioni delle correlazioni sono prodotte da cambiamenti nella sensibilità dei rendimenti agli shock macroeconomici e di mercato, oppure se si tratta semplicemente di un effetto "meccanico" legato a variazioni repentine della volatilità di mercato. Un'osservazione superficiale potrebbe anche impiegare questa semplice evidenza come sostegno dell'ipotesi che i mercati siano diventati ancora più integrati globalmente, ma anche su questo sarebbe necessario impiegare metodi quantitativi più sofisticati.

Resta comunque evidente l'aumento delle correlazioni dell'ultimo quindicennio. Puro aumento della volatilità o maggiore interdipendenza tra i mercati, si tratta di un fenomeno dai risvolti molto significativi, non ultimi quelli sul potenziale di diversificazione degli investimenti. Nel primo caso, che al momento sembra essere confermato dai risultati preliminari del nostro lavoro, sorgerebbe spontanea la domanda: cosa c'è a monte di un aumento così massiccio della turbolenza sui mercati?

Credo che una possibile risposta potrebbe risiedere nei ruoli innovativi che la regolazione finanziaria e la politica monetaria hanno cominciato a giocare proprio alla fine degli anni 1990. Deregolamentazione e forte impulso alla liquidità straordinaria da parte delle banche centrali potrebbero aver influenzato la propensione alla volatilità dei mercati azionari. Si tratta di un'ipotesi difficile da verificare, ma estremamente interessante...

11 febbraio 2012

I guastafeste, parte prima


Mentre i giornali italiani continuano a celebrare i fasti del viaggio del premier in USA e la recente compressione degli spread italiani, la realtà ostinatamente continua a presentarsi alla nostra porta.

S&P ha appena declassato il rating di 34 delle 37 istituzioni finanziarie italiane che segue, motivando la decisione con il continuo deterioramento delle prospettive diredditività e di rischio delle banche, fortemente condizionate dalle "zavorre" della situazione macroeconomica del Paese.

La società ha anche collocato l'industria bancaria del Paese all'interno di un sistema di classificazione, con punteggio da 1 a 10, in funzione dell'esposizione al rischio. L'Italia adesso figura nella classe 4, insieme ai sistemi bancari di Repubblica Ceca, Sud Africa, Messico e Brasile. Ogni commento, in linea con le nostre ben note perplessità sul sistema bancario nostrano, è abbondantemente superfluo...

In basso alleghiamo l'interessantissima "motivazione" di questa sentenza, con in grassetto i passaggi che mi sembrano più evocativi. I più istruttivi sono senz'altro i seguenti:

"Italy has a large and diversified economy with a sizable domestic market and manufacturing export sector. However, we anticipate weaker GDP growth for Italy relative to its peers owing to Italy's high public sector debt, which in our view is discouraging investment; weakening Italy's export competitiveness; and to Italy's worsening demographic profile."

"While these profitability ratios would be only slightly below the average for 2009-2010, they would represent less than half of returns in 2005-2008. Persistently weak profitability for the next few years, together with increased cost of capital for the whole banking industry, could lead Italian banks to write down a large part of the goodwill they booked during the consolidation phase of the past decade. We note, though, that Italian banks' risk appetite generally remains restrained, reflecting their traditional business model."



LONDON (Standard & Poor's) Feb. 10, 2012--Standard & Poor's Ratings Services today said it has lowered its ratings on 34 Italy-based financial institutions. The downgrades follow the lowering of the unsolicited long- and short-term sovereign credit ratings on the Republic of Italy (BBB+/Negative/A-2; see "Italy's Unsolicited Ratings Lowered To 'BBB+/A-2'; Outlook Negative," published Jan. 13, 2012, on RatingsDirect on the Global Credit Portal). They also reflect the revision of our Banking Industry Country Risk Assessment (BICRA) on Italy to group '4' from group '3', and of our economic risk and industry risk scores--both components of the BICRA--on Italy to '4' from '3' (see "BICRA On Italy Revised To Group '4' From Group '3' On Weakening Economic And Banking Industry Conditions," published Feb. 10, 2012).
In addition, we have affirmed our ratings on two Italian financial institutions and removed them from CreditWatch with negative implications. We have also kept the ratings on one Italian financial institution on CreditWatch with negative implications.
Mainly Negative Rating Actions Taken On 37 Italian Financial Institutions On Sovereign Downgrade And BICRA Change

OVERVIEW
  1. On Jan. 13, 2012, Standard & Poor's lowered its unsolicited long- and short-term sovereign credit ratings on the Republic of Italy to 'BBB+/A-2' from 'A/A-1', assigned a negative outlook, and removed the ratings from CreditWatch negative.
  2. In our view, Italy's vulnerability to external financing risks has increased, given its high external public debt, resulting in Italian banks' significantly diminished ability to roll over their wholesale debt.
  3. We anticipate persistently weak profitability for Italian banks in the next few years, and a risk-adjusted return on core banking products that may not be sufficient for banks to meet their cost of capital. We believe this may be negative for the Italian banking industry's stability.
  4. We are revising our Banking Industry Country Risk Assessment (BICRA) on the Republic of Italy to group '4' from group '3', and our economic risk and industry risk scores, two components of the BICRA, to '4' from '3'.
  5. Our revised BICRA and economic risk score for Italy have resulted in primarily negative rating actions for the Italian banks we rate.

BICRA ACTION
On Feb. 10, 2012, Standard & Poor's Ratings Services revised its Banking  Industry Country Risk Assessment (BICRA) on the Republic of Italy (unsolicited ratings, BBB+/Negative/A-2) to group '4' from group '3'. It has also revised  the economic risk and industry risk scores, two components of the BICRA, to '4' from '3'. These revisions follow our recent downgrade of the Republic of Italy (see "Italy's Unsolicited Ratings Lowered To 'BBB+/A-2'; Outlook Negative ," published Jan. 13, 2012, on RatingsDirect on the Global Credit Portal).
RATIONALE
The BICRA change reflects our view that Italy's vulnerability to external financing risks has increased, given its high absolute amount of external public debt. This results in adverse spill-over effects on Italian banks, in particular significantly diminishing their ability to roll over their wholesale debt. In addition, we anticipate persistently weak profitability for Italian banks in the next few years, and a risk-adjusted return on core banking products that may not be sufficient for banks to meet their cost of capital. We believe this may be negative for the Italian banking industry's stability.
Standard & Poor's BICRA rankings integrate its view of the strengths and weaknesses of a country's banking system compared with those of other countries. A BICRA is scored on a scale from 1 to 10, ranging from the lowest-risk banking systems (group '1') to the highest-risk (group '10'). Countries in BICRA group '4' include Czech Republic, South Africa, Mexico and Brazil.
Our revised BICRA and economic risk score for Italy have resulted in primarily negative rating actions on the Italian banks we rate (see "Mainly Negative Rating Actions Taken On 37 Italian Financial Institutions On Sovereign Downgrade And BICRA Change," published Feb. 10, 2012).
The economic risk score of '4' is based on our revised opinion of Italy's "economic imbalances," one of the score's main components,  to "intermediate risk" from "low risk," as our criteria define these terms.
The economic risk score is an input in our calculation of risk-weighted assets (RWA) for banks under our risk-adjusted capital (RAC) framework (see "Bank Capital Methodology And Assumptions," published Dec. 6, 2010). Our revised economic risk score will likely lead to moderate declines in our RAC ratios for banks operating in Italy.
We believe that Italy's higher vulnerability to external financing may represent a risk to the sustainability of Italy's balance of payments. The deepening political, financial, and monetary problems in the Economic and Monetary Union (EMU or the eurozone) are exacerbating the external funding constraints on the Italian public and private sectors, in our opinion. Italy's external financing costs have risen markedly in recent months and may remain elevated for an extended period as the Italian government and banks scale back cross-border financing. Still, we acknowledge that imbalances in asset prices and the buildup of leverage are low in Italy. In our view, these factors remain key supports for Italian banks' creditworthiness.
Our assessment of "intermediate risk" for "economic resilience" and "credit risk in the economy," the two remaining components of the economic risk score, remains unchanged.
Italy has a large and diversified economy with a sizable domestic market and manufacturing export sector. However, we anticipate weaker GDP growth for Italy relative to its peers owing to Italy's high public sector debt, which in our view is discouraging investment; weakening Italy's export competitiveness; and to Italy's worsening demographic profile.
The Italian private sector carries, in our view, moderate debt at 128% of GDP expected at year-end 2011. Household indebtedness, at 45% of GDP, is low relative to levels in peer countries. In addition, Italian households' financial wealth is about two times GDP, which sustains their creditworthiness in case of difficulties. Italy's banking sector typically has high levels of nonperforming assets (NPAs) relative to countries that we assess as having similar economic risk. The high NPAs reflect Italian banks' large share of loans to small and midsize enterprises, which form the backbone of the Italian economy. Extensive use of collateral has, however, generally enabled banks to post only moderate credit losses.
The industry risk score of '4' is based on our revised opinion of the Italian banking sector's "competitive dynamics" and "systemwide funding" to "intermediate risk" from "low risk," as our criteria define these terms.
Our view that Italian banks' profitability will likely remain weak in the next few years reflects Italy's flat economic growth, and banks' increased cost of funding and still high credit provisions. Risk-adjusted returns on core banking products may not be sufficient for banks to meet their cost of capital. We believe this may be negative for the Italian banking sector's stability. Based on our estimates, the return on adjusted assets related to domestic banking business should average close to 30 basis points in 2011-2013. The net income-to-revenue ratio should bottom out in 2012 and remain below 10% in the same period. While these profitability ratios would be only slightly below the average for 2009-2010, they would represent less than half of returns in 2005-2008. Persistently weak profitability for the next few years, together with increased cost of capital for the whole banking industry, could lead Italian banks to write down a large part of the goodwill they booked during the consolidation phase of the past decade. We note, though, that Italian banks' risk appetite generally remains restrained, reflecting their traditional business model.
We believe that "systemwide funding" risk for Italian banks has increased, mainly owing to the effects of the sovereign debt crisis in Europe. European Central Bank (ECB; AAA/Stable/A-1+) funding provided to Italian banks was up sharply at €203 billion at end-January 2012, or 6% of total funding, versus €41 billion at end-June 2011. In our view, the steep rise is due to Italian banks having largely lost their access to external wholesale debt capital markets, which we believe will in turn significantly reduce their ability to roll over their wholesale debt this year. In our opinion, Italian banks' recourse to the ECB may further increase after the ECB's next Long-Term Refinancing Operation (LTRO) scheduled in late February 2012 (for further details on ECB funding to eurozone banks, see "Assessing The Severity Of The Eurozone Recession Is A Close Call," published Jan. 31, 2012). We believe that the ECB's proactive stance on providing liquidity to Italian banks has avoided a domestic credit crunch.
Italian banks' recourse to wholesale domestic or cross-border funding is contained, in our opinion, and to a degree offsets our concerns about systemwide funding. The banking system's core customer deposits (including bonds placed with retail clients) cover between 90% and 100% of its customer loans. Italian banks depend on external debt to cover about 15% of total loans on a net basis, by our estimates. But a significant portion of external debt relates to funding that foreign banks have provided to domestic subsidiaries, which we believe is clearly less confidence sensitive. We also believe that Italian banks' pools of assets eligible for ECB funding is sufficiently wide to cover 2012 wholesale maturities. In addition, the Italian government has offered Italian banks the option to request a state guarantee for ECB funding received in an amount up to their regulatory capital. The Bank of Italy has also temporarily expanded the eligibility criteria for credit claims used as collateral in the ECB's lending operations, as part of the ECB's support measures for banks. As for other eurozone members, our assessment of Italy's systemwide funding incorporates our expectation that the ECB will maintain its liquidity support for banks.
We have maintained our assessment of "intermediate risk" for Italy's "institutional framework." Italy has gradually aligned its regulatory standards with international best practices in recent years. Supervision is effective and hands-on, in our opinion. Italy's banking regulator has a good record in dealing with individual bank crises, in our view. However, we believe that some banks, particularly small ones, tend to have less-than-optimal governance. Still, Italy's banks still generally have some leeway that allows them to delay recognizing credit losses.
We classify the Italian government as "supportive" toward its banking sector. We recognize the government's record of providing support to the banking system in times of stress.

6 febbraio 2012

Il debito fa male

Insieme a Simone Salotti (National University of Ireland, Galway) sto ultimando una ricerca sull'effetto del debito pubblico su investimenti aggregati e produttività. L'analisi empirica studia 20 paesi OECD dal 1970 ad oggi, e cerca di approfondire i termini della ben nota affermazione di Reinhart e Rogoff relativa all'associazione negativa tra debito ecrescita economica.

I risultati econometrici sono molto chiari: un debito pubblico più elevato causa minore spesa per investimenti, ossia una più lenta accumulazione di capitale, che è invece essenziale per la crescita di lungo termine dell'economia. Inoltre, debito maggiore significa anche più lenta crescita della produttività, misurata come GDP per ora lavorata, una misura standard delle possibilità produttive. I risultati sono molti forti, e frutto di verifiche molto attente a evitare errori di identificazione di una relazione causale inversa. Quindi, siamo certi di aver identificato una relazione che va dall'alto debito alla bassa crescita, e non viceversa.

I risultati saranno presto disponibili in un working paper, ma nel frattempo il grafico in basso (click per ingrandire) offre una prospettiva molto chiara.



La figura mostra il tasso medio di crescita degli investimenti e il tasso medio di crescita della produttività per livelli diversi (30%, 60% e 90%) del rapporto debito pubblico/PIL nei 20 paesi del campione di 40 anni. Le soglie sono state scelte per analogia con quelle del famoso studio di Reinhart e Rogoff. Dal grafico emerge che più elevati livelli di debito sono chiaramente associati con tassi di crescita più bassi sia di accumulazione di capitale che di produttività.

1 febbraio 2012

A long bond

Una visione di lungo termine è una delle risorse più importanti della vita. Specialmente in economia. E negli investimenti.

Il grafico in basso relativo al tasso di interesse a lunga scadenza negli USA (click per ingrandire) aiuta a capire alcune centinaia di fenomeni, tra i quali la profondità delle trasformazioni nel sistema monetario che stiamo vivendo oggi...