Loading...

Follow Investors Corner - BNP Paribas Investment Partners on Feedspot

Continue with Google
Continue with Facebook
or

Valid

Jane Ambachtsheer, Global Head of Sustainability at BNP Paribas Asset Management, addresses some of the main issues around sustainable investing and the use of environmental, social and governance (ESG) criteria in selecting assets for investment that are appropriate for a future low-carbon, inclusive world.

What does sustainable investing mean to you and what is the focus and approach of BNPP AM?

For us, sustainable investing includes these elements:

  • One is responsible corporate behaviour. This is aligned with the UN Global Compact. It involves sector policies which rule out investments in, for example, coal and tobacco.
  • Another component is ensuring that each of our around 80 investment process fully integrates our ESG guidelines.
  • On stewardship, we actively exercise our proxy votes. We are not shy about voting against management when we believe that it needs to focus more on disclosure around climate, for example, or diversity issues.
  • Finally, we focus on the shift towards a low-carbon, inclusive economy and the roles individual companies and countries are taking.
Exhibit 1: Three challenging areas of action

Source: Global Sustainability Strategy by BNP Paribas Asset Management; March 2019

On stewardship, what is happening in terms of a broader push?

BNP Paribas Asset Management is an active member of Climate Action 100+ and one of 300 or so investors representing USD 30 trillion who are targeting the biggest greenhouse gas emitters. We are asking these companies to have a business plan aligned with keeping the rise in global temperatures to 2C; to have climate-competent boards; and to provide appropriate disclosure on climate-related issues.

There is a fundamental shift in where money is going as awareness grows, asset owners become more aligned and are working to clarify and harmonise the terminology and disclosure. What is BNPP AM’s contribution?

We support GRASFI. This alliance brings together 18 top universities as they seek to establish an academic discipline focused on sustainable finance and investment.

We are also a member of the CFA committee establishing a new ESG standard. In our own reporting, we are covering the impact of each of the elements of sustainable investing: exclusions, risk management, engagement and advocating a sustainable economic growth model. We are going to publish, for the first time, a robust report on all our stewardship activity.

Has sustainable investing become mainstream?

We have flipped from a place where managers needed to explain why they were putting such focus on sustainable investing to a point now where managers would need to justify why they’re not putting more focus on it. It has become a must-have.

This means that as asset owners, we must be active stewards, think about the future, and protect the economy over the long term. This is the right thing to do in terms of financial results, as well as socially and environmentally.

Exhibit 2: BNP Paribas Asset Management’s investment beliefs

Source: Global Sustainability Strategy by BNP Paribas Asset Management; March 2019

Governance has become a prerequisite as an investment criterion, also in emerging markets. It has become a topic for issuers such as superstar firms, particularly in the tech sector where governance and shareholder access are often not as good as they are in other sectors. More and more investors are engaging on governance.

What is next in terms of asset classes?

After equities, sustainable investing is entering areas which no one has really explored from a sustainability perspective.

  • For example, private debt investing. Here it is more difficult to get information from the companies, you have a narrow window to do your analysis, and you may not have direct access to the company.
  • In structured products, we are working on a responsible derivatives policy.
  • In fixed income, we are finalising our approach to researching sovereigns. We focus on the ESG performance of a country: not just in absolute terms, but also in terms of the relative ability of that country to actually implement strong ESG practices in the context of their income levels, whether it is social infrastructure, schools, hospitals, etc. We also look at a country’s performance on climate. Are countries on a 5°C pathway or a 2°C pathway?
Climate change is central when it comes to sustainability. How does an asset owner approach this?

Clients demand and deserve to work with financial institutions which acknowledge and embrace their role in influencing a future aligned with the Paris Agreement. We tackle that through Climate Action 100+ and we are actively encouraging policymakers to implement the changes that are required to put the world on a low-carbon pathway.

We are also working to progressively align our portfolios with the International Energy Agency’s 2°C scenario and steps towards decarbonisation.

How can investors select the right funds and managers?

You need to look at both the investment process as well as the portfolio. For BNPP AM, the process involves our ESG integration guidelines: we are making sure that each investment process meets the same standards and that we can communicate on that. From my experience in the industry, that is unique.

We want all of our portfolios to have a better ESG score than the relative benchmark. To help investors in their evaluation of these strategies, we want to be clear on what we are doing, why we are doing it and how we are doing it and provide consistent reporting on the actual results.

What is the biggest challenge you face in this area? And the biggest opportunity?

Getting consistent and reliable information is a challenge as it is not always reported by the company or the issuer. Half of the carbon foot printing data now is estimated. Helpfully, there is a push on corporate disclosure from investors and regulators.

To me, the big opportunity is in the role finance can play in establishing a low-carbon, inclusive economy. For the financial community, that is not just an opportunity, but also an obligation. We are not just bystanders.

For more articles on sustainable investing, click here >

Discover the sustainable investment solutions of BNP Paribas Asset Management and read more about what sustainability means to us.


Read more about our Global Sustainability Strategy

 

Writen by Jane Ambachtsheer. The post Sustainable investing: we are not just bystanders appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

Reliance on coal for the generation of electricity varies greatly across Europe. While some countries have almost eliminated coal usage, others remain heavily reliant. The interactive map below shows the percentage of electricity production from coal in each of the member states of the European Union.

For more articles on sustinable investing, click here >

 

Writen by Investors' Corner Team. The post Coal usage in the European Union appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

The Bank of England’s stance on both monetary and fiscal policy, as well as the state of the UK economy and the attractiveness of UK assets, all hinge on the outcome of the Brexit process.

The political situation regarding Brexit remains highly fluid. Predicting the end result is particularly complicated due to the multitude of plausible outcomes. At the time of writing (March 2019), Parliament appears to have moved to actively prevent a disorderly, no-deal, Brexit. Theresa May’s government is working on a cross-party alternative to her withdrawal deal, boosting the chances of a softer Brexit. But while the tail risk of a no-deal, cliff-edge exit has decreased substantially, the risk of a UK general election being held in the short term has increased.

As to what might happen – some scenarios 1. The withdrawal agreement goes through

Should politicians reach a cross-party compromise allowing Theresa May’s agreement to pass, the transition period in which the UK remains in the single market kicks in. Negotiations on the future relationship between the UK and the EU continue.

Such an outcome would probably provide the market some short-term relief. But the can would simply be kicked down the road for another 21 months (or longer) as details of future trade arrangements will need to be hashed out. It is far from clear whether the UK economy will enjoy a bounce.

Of course, avoiding a disruptive no-deal exit should boost investor sentiment. The British pound would likely enjoy a relief rally.

Consumer spending has not yet been meaningfully affected by worries over Brexit. If anything, stockpiling in preparation for Brexit disruptions has provided a temporary boost to the UK’s manufacturing PMI in recent months.

And should the transition period go live, the Bank of England (BoE) will likely follow through with its hawkish rhetoric and the market could quickly reprice front-dated UK interest rates to reflect faster rate rises in the next couple of years.

Breakeven inflation (BEI) rate spreads, particularly those at the shorter end of the curve, will likely tighten as the market will no longer need to worry about higher imported prices and/or trade tariffs in the near future.

2. No-deal Brexit

While a no-deal Brexit looks less likely, having a political consensus against a no-deal is not enough – there is still no consensus around an alternative course of action. And the risk of a no-deal Brexit could re-emerge through a second referendum. Hence, a no-deal outcome cannot be written off yet.

Should there be a no-deal exit, the UK will likely fall into the World Trade Organization’s (WTO) most favoured nation’s tariff regime. But to trade under WTO law, the UK will have to establish its own schedules, which will then have to be accepted by all WTO members. The UK will also face non-tariff barriers such as EU regulation checks at the borders and trade quotas.

The financial services industry would face significant regulatory limitations and disruptions. In the BoE’s analysis, the economic hit from a no-deal would range from -4.75% to -7.75% in five years’ time.

Sterling would likely take a hit in the short term. Currency weakness, and perhaps also a shortage of goods, would drive inflation higher, again at least in the short term. This in turn will likely push shorter-dated BEI rates higher. Gilts would likely rally as the market prices out future rate increases and instead prices in potential rate rises or even a resumption of quantitative easing (QE).

3. A second referendum

Increasing numbers of MPs appear to have concluded that it is better to ask the people to make an incredibly difficult decision for them. However, the path to the second referendum is uncertain. It would be unlikely to be a repeat of the first one, so the phrasing of the questions on the ballot will not be easy.

A decision to hold a second referendum will probably be market-positive initially as opinion polls suggest that Remain has a better chance of winning this time. However, any positive market reaction will likely be offset by the prolonged uncertainty during the campaign period and the danger that a second referendum might lead to a no-deal Brexit. If the public chose the no-deal option, there would be little time or patience for the EU to continue Brexit discussions.

4. General election

The risk of another vote of no confidence in the government in an attempt to trigger a general election has increased, and given Theresa May’s cross-party attempt to break the deadlock, Conservative Brexiteers would likely be more rebellious. As support for the Conservative party has weakened meaningfully in recent polls, markets may conclude that the probability of a Corbyn-led Labour government is rising.

Labour’s policy agenda, which favours higher income and corporate taxes to pay for higher government spending, renationalisation of utility and rail companies, abolishing tuition fees, limiting the rise in pension age, etc., is generally perceived to be fiscally expansionary and detrimental to gilts and sterling.

That said, we believe market concerns over a Corbyn-led government should be fading. Recent election polls are pointing to greater political fragmentation, making it harder for an eventual coalition government to hammer out compromises on major policy items.

Taking a tactical position in the UK

Given the low visibility on the economic outlook and continued domestic political risks, our active strategies remain focused on near-term tactical opportunities. In duration, Brexit uncertainty, concerns over global growth, the recent BoE QE reinvestment flows, as well as duration demand on the back of the UK inflation linked bond index extension, pushed 30-year real yields to a low of -2% and 10-year conventional yields to just below 1% in late March.

With global growth sentiment recovering after the recent bounce in China’s PMI data, PM May’s moving towards a cross-party solution to break the political deadlock, and the conclusion of the BoE’s reinvestment programme and the index extension, we believe a short duration position at current levels offers some tactical value.

This is an extract from the Q1 2019 Inflation-Linked Bonds Outlook published in March. To read the full version, click here >

For more articles by Cedric Scholtes, click here >

For more articles by Jenny Yiu, click here >

 

Writen by Cedric Scholtes. The post UK index-linked gilts market: outlook and exposures appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

External factors such as uncertainty over the Brexit process appear to be leading some investors to doubt there will be any meaningful re-acceleration of growth in the eurozone, although recent ‘soft data’ suggests consumption is resilient.

  • Inflation is sluggish
  • BEI rate plummeted between November and March
  • Various factors could limit the potential rise in core government bond yields

Over the first quarter of 2019, we held the view that the slowdown in eurozone economic growth was largely due to external weakness, and that concerns over further external shocks should gradually ease given the dovish stance by major central banks around the world, along with policy easing measures by the Chinese authorities and the de-escalation in US-China trade tensions.

At the same time, we believed that favourable financial conditions, further improvement in employment and wage growth, as well as a more expansionary fiscal stance in major eurozone member states, should continue to support private consumption in the eurozone.

With our base case assumption that the eurozone economy should grow at slightly above trend, albeit below 2017 levels, and our expectation that worst-case concerns over an economic slowdown would not materialise, our inflation-linked bond portfolio was positioned with a modest duration underweight and a long breakeven inflation (BEI) rate position to capture a potential recovery in risk sentiment.

Eurozone inflation developments: still lacklustre

However, despite some early signs of recovery in eurozone economic activity, the extent of the improvement in Q1 was too modest to reassure many investors on any meaningful economic reacceleration given the lingering risks from external factors and Brexit-related uncertainty.

At the same time, inflation developments are lacklustre. Revisions to Germany’s inflation data due to a methodology change led to an unfavourable one-off impact on the unrevised eurozone Harmonised Index of Consumer Prices (HICP) ex-tobacco to which inflation-linked bonds are referenced. Our eurozone active positions suffered as a result and detracted slightly from performance.

Looking ahead, recent incoming data has reinforced our view that we have probably already seen the trough in global growth. Purchasing managers’ index (PMI) surveys are showing early signs of a bounce-back in Chinese growth, and we expect the easing measures announced by the Chinese authorities to continue to pass through to the real economy and keep growth moving in the right direction.

In the eurozone, while the manufacturing sector is slow to recover, there are signs of more green shoots across member states. Recent eurozone services PMI surveys have confirmed our view that private consumption is resilient.

BEI rate: historically large fall

In breakeven inflation, having spent much of 2018 gyrating within the 1.6% to 1.8% range, the 5-year/5-year forward BEI rate started to fall precipitously last November to a low of around 1.3% in March. The drop is large by historical standards, and reflects the deceleration in both global and eurozone growth, expectations that eurozone inflation will remain low, and scepticism over the ECB’s ability to deliver on its inflation target given its constrained policy tools.

The inflation picture will likely be murky in the near term as volatile seasonal changes in holiday package prices around the Easter holiday will make inflation data difficult to read. The recent growth worries may also slow the process of higher wages passing through to inflation. Our near-term view is that eurozone BEI rates offer some tactical value, as we remain optimistic that economic activity will recover in the coming months.

We see the ECB’s discussion about “tiering” as a sign of preparing for potential additional easing should economic growth disappoint further. Such action may relieve investor concerns about the ECB running out of ammunition.

Adverse scenarios may be priced in, longer term

We expect the combination of economic green shoots, an ECB more prepared to ease, and positive inflation carry seasonals to help BEI rates recover in the near term. Over the longer term, however, if the growth momentum continues to stall and if the disinflationary shock intensifies, we see increasing risks of the market pricing in a higher probability of adverse scenarios, such as deflation, given that central bank policy rates will likely still be close to the “effective lower bound” and given the political hurdles to returning to sizable quantitative easing programmes.

In terms of duration, we held a modest tactical underweight in duration early in the year as we saw asymmetric risk-reward tipped in favour of higher yields in front of the wall of government supply scheduled for January.

However, we were quick to exit the position, at a small loss, after we saw that the market absorbed the new issuance well and German Bund yields continued to hover at low levels given the lack of economic recovery.

Various factors could limit rise in core yields

More recently, we have started to re-establish a modest duration underweight in core eurozone countries, motivated by the concern over a manufacturing slump and external weakness dragging domestic activity down dissipating, and the apparently diminishing probability of a disorderly hard Brexit.

Over the longer term, however, we see

  • the backdrop of low inflation
  • lingering threats of US tariffs on German car exports
  • unresolved political and fiscal issues in Italy.

This should limit the extent of a potential rise in core government bond yields.

In core yield curves, we have gradually moved our inflation-linked bond holdings from the intermediate sector to the shorter end of the real yield curve to capture the turn to more favourable inflation carry seasonals in the coming couple of months.

In ‘peripheral’ yield curves, we held a modest flattening exposure to express our defensive view. In Italy, while some of the major risk events such as the rating agency decisions and the budget fight are behind us, we expect the bond market to continue to be clouded by uncertainty and volatility.

Recent macroeconomic developments are not encouraging – real GDP growth has slowed; inflation has remained subdued, and consumer confidence remains soft. Italy’s low potential trend growth and the related structural weakness remained unresolved. We also see a disconnect between the optimism of the new European Parliament being more friendly in relation to Italy’s fiscal reality and see a high likelihood of volatility returning to the Italian government bond market.

This is an extract from the Q1 2019 Inflation-Linked Bonds Outlook published in March. To read the full version, click here >

For more articles by Cedric Scholtes, click here >

For more articles by Jenny Yiu, click here >

For more articles on fixed income, click here >

 

Writen by Cedric Scholtes. The post Eurozone inflation-linked bond market: outlook and exposures appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

A major quant seminar in the UK brought out the latest academic thinking on the advantages – and dangers – of using new data techniques like machine learning, AI and text analysis to improve investment outcomes

  • Machine learning – useful for FX, commodities and finding patterns in less-invested asset classes
  • Backtesting in the machine learning era needs to be honest, precise and scientific
  • Finding value from financial text

The recent Inquire Europe and Inquire UK joint Spring 2019 seminar in Windsor, UK, included presentations and discussion of a broad range of topical investment-related issues.

Machine learning and AI

Sandy Rattray, CIO at Man Group, viewed machine learning and artificial intelligence as the biggest fad in 2018, but he believes machine learning techniques are not particularly useful for forecasting asset returns at horizons of six months and longer.

Where Sandy sees useful applications is in trading, in particular FX, where Man Group collects most big data, and in commodity markets. He also sees it as useful for text mining and for discovering patterns in less exploited asset classes.

In terms of new trends in quantitative investing, Sandy highlighted credit multi-factor as something to watch. He also wants to increase interaction between quantitative and fundamentals fund managers but perceives the poor quantitative skills of some fundamental managers as a difficulty to overcome. The fact that managing concentrated portfolios is difficult for quantitative managers does not help either, in his view.

When it comes to new data, Sandy sees many people creating new datasets for fundamental managers, but he is unsure of their value, in particular taking into account their (often high) cost and the trend for lower management fees, even if he certainly does not think data budgets are likely to go down.

Finally, when it comes to long-term investing, he believes that there are not many factors adding value. He advises investors to stick with those they know.

A backtesting protocol in the era of machine learning

Campbell Harvey, Duke’s Fuqua School of Business presented his recent paper, co-authored with Harry Markowitz and Rob Arnott, A Backtesting Protocol in The Era Of Machine Learning. After numerous anecdotal examples of the dangers of blindly applying machine learning techniques to asset return forecasting, Campbell laid out the key attention points defended in their paper.

The first is to avoid HARKing (Hypothesizing After the Results are Known). Data-mined factors should be treated with much more scepticism than a factor from economic theory. The second is to be aware of the multiple testing problem, i.e. that when we run a hypothesis test there is only a small chance of finding a bogus significant result but if we run thousands of tests, typical of machine learning, then the number of false alarms increases dramatically. Researchers should not stop researching as soon as they find a good model. All variables set out in the research agenda should be investigated.

The third is to be aware of data integrity, to what extent outlier exclusions and data transformations were set in advance and make sense, and to what extent results are resilient to minor changes in transformations.

The fourth is to be honest with cross-validation, avoiding later modifying an in-sample model to fit out-of-sample data, making sure that the out-of-sample analysis is representative of live trading, and that realistic transactions costs were accounted for.

The fifth is awareness of model dynamics – making sure the model is resilient to structural changes and that steps were taken to minimise overfitting of the model dynamics, the tweaking of the live model and the risk of overcrowding.

The sixth is to avoid complexity, i.e. that models avoid the curse of dimensionality, that the simplest practicable model specification is retained, and that results from machine learning can be interpreted rather than used as black box. Regularisation, i.e. introducing constraints to achieve model simplification that prevents overfitting is good practice.

Finally, the seventh is the need to have a scientific research culture, i.e. to reward quality rather than just the finding of a positive back-test of a strategy.

A good illustration of how violating these principles can lead dangerous results based on overfitting was that of a strategy that invests in an equally weighted portfolio of stocks with “S” as the third letter of their ticker symbol and shorts an equally weighted portfolio of stocks with “U” as the third letter of their ticker symbols. This strategy, cited in their paper, was found after trying thousands of combinations. The strategy would have generated a significantly high risk-adjusted return when applied to US stocks between Jan-63 and Dec-15, even in the Global Financial Crisis of 2008, and with a low turnover of less than 10% per year.

Where’s the value in unstructured data?

Steven Young from Lancaster University gave a tutorial on the value of unstructured data for investing. The motivation for using and modelling with text is simple: many facts are not easily translated into summary numbers, and nuances are better expressed in text.

Steve’s objective was to propose a framework for using financial text and thinking about the sources of value associated with text. Clearly, even before starting, researchers should ask themselves about the comparative advantage of their research. Text mining packages such as those found in R, Python or SAS mean that the low hanging fruit has already been found, picked and eaten.

For example, when it comes to text analysis of 10-K and 10-Q reports for US firms, more than 60 papers have been produced. Once the added value of the project is clear, Steve proposes a four-step framework:

  1. Corpus Creation, corresponding to the Definition of the Problem
  2. Cleaning & Pre-Processing
  3. Annotation
  4. Processing, corresponding to the Search of Meaning.

For the Corpus Creation, the three broad genres of finance-related textual content are

  1. forum, blogs and wikis
  2. ii) news and research reports
  3. iii) content generated by firms.

Analysing multiple genres to contrast content views and triangulate results is likely to add more value. It is also important to decide whether to use the entire text or just part of it. Sharper conclusions may be possible by focusing on particular sections. Then one must harvest the textual data, clean it and pre-process it removing unwanted contents and organising the relevant unstructured text into structured text or numerical data arranged in tables.

The goal is to construct a term document matrix, i.e. TDM. Pre-processing may also include the removal of punctuation and numbers, removal of stop words, stemming, and disambiguation. Annotation may be manual or automatic and is critical for disambiguation and for feature extraction.

Manual tagging is more likely subjective but can play an important role for training in Big Text applications. In turn, automated tagging may use a number of available resources for Part-Of-Speech (POS), for morphology, grammar and syntax, for semantics and for pragmatic annotation.

Finally, processing may be based on simple word frequency counts using general dictionaries such as DICTION, General Inquirer, LIWC and others, or based on more refined approaches based on domain-specific lexicons such as Netlingo or Provalis.

There are many possible refinements in processing. Examples include handling categorisation, influence, emphasis words, specificity, similarity, obfuscation and fake news. Another axis often explored is weighting, according more importance to an unusual word or to words more closely related to the underlying construct.

The number of research papers on the value of text for investing keeps growing. Besides firm sentiment, some examples of successfully identifying firm- and manager-level risk factors include fraud and misreporting, CEO personality traits, idiosyncratic political risk, firm geographic exposure and financial constraints.

In Steven’s view, it is clear that automated text mining is a tool of increasing importance for quantitative investment.

For more articles by Raul Leote de Carvalho, click here >

 

Writen by Raul Leote de Carvalho. The post New data, new methods: highlights from the Inquire quant seminar – Part 1 appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

A major quant seminar in the UK brought out the latest academic thinking on the advantages – and dangers – of using new data techniques like machine learning, AI and text analysis to improve investment outcomes

  • Machine learning – useful for FX, commodities and finding patterns in less-invested asset classes
  • Backtesting in the machine learning era needs to be honest, precise and scientific
  • Finding value from financial text

The recent Inquire Europe and Inquire UK joint Spring 2019 seminar in Windsor, UK, included presentations and discussion of a broad range of topical investment-related issues.

Machine learning and AI

Sandy Rattray, CIO at Man Group, viewed machine learning and artificial intelligence as the biggest fad in 2018, but he believes machine learning techniques are not particularly useful for forecasting asset returns at horizons of six months and longer.

Where Sandy sees useful applications is in trading, in particular FX, where Man Group collects most big data, and in commodity markets. He also sees it as useful for text mining and for discovering patterns in less exploited asset classes.

In terms of new trends in quantitative investing, Sandy highlighted credit multi-factor as something to watch. He also wants to increase interaction between quantitative and fundamentals fund managers but perceives the poor quantitative skills of some fundamental managers as a difficulty to overcome. The fact that managing concentrated portfolios is difficult for quantitative managers does not help either, in his view.

When it comes to new data, Sandy sees many people creating new datasets for fundamental managers, but he is unsure of their value, in particular taking into account their (often high) cost and the trend for lower management fees, even if he certainly does not think data budgets are likely to go down.

Finally, when it comes to long-term investing, he believes that there are not many factors adding value. He advises investors to stick with those they know.

A backtesting protocol in the era of machine learning

Campbell Harvey, Duke’s Fuqua School of Business presented his recent paper, co-authored with Harry Markowitz and Rob Arnott, A Backtesting Protocol in The Era Of Machine Learning. After numerous anecdotal examples of the dangers of blindly applying machine learning techniques to asset return forecasting, Campbell laid out the key attention points defended in their paper.

The first is to avoid HARKing (Hypothesizing After the Results are Known). Data-mined factors should be treated with much more scepticism than a factor from economic theory. The second is to be aware of the multiple testing problem, i.e. that when we run a hypothesis test there is only a small chance of finding a bogus significant result but if we run thousands of tests, typical of machine learning, then the number of false alarms increases dramatically. Researchers should not stop researching as soon as they find a good model. All variables set out in the research agenda should be investigated.

The third is to be aware of data integrity, to what extent outlier exclusions and data transformations were set in advance and make sense, and to what extent results are resilient to minor changes in transformations.

The fourth is to be honest with cross-validation, avoiding later modifying an in-sample model to fit out-of-sample data, making sure that the out-of-sample analysis is representative of live trading, and that realistic transactions costs were accounted for.

The fifth is awareness of model dynamics – making sure the model is resilient to structural changes and that steps were taken to minimise overfitting of the model dynamics, the tweaking of the live model and the risk of overcrowding.

The sixth is to avoid complexity, i.e. that models avoid the curse of dimensionality, that the simplest practicable model specification is retained, and that results from machine learning can be interpreted rather than used as black box. Regularisation, i.e. introducing constraints to achieve model simplification that prevents overfitting is good practice.

Finally, the seventh is the need to have a scientific research culture, i.e. to reward quality rather than just the finding of a positive back-test of a strategy.

A good illustration of how violating these principles can lead dangerous results based on overfitting was that of a strategy that invests in an equally weighted portfolio of stocks with “S” as the third letter of their ticker symbol and shorts an equally weighted portfolio of stocks with “U” as the third letter of their ticker symbols. This strategy, cited in their paper, was found after trying thousands of combinations. The strategy would have generated a significantly high risk-adjusted return when applied to US stocks between Jan-63 and Dec-15, even in the Global Financial Crisis of 2008, and with a low turnover of less than 10% per year.

Where’s the value in unstructured data?

Steven Young from Lancaster University gave a tutorial on the value of unstructured data for investing. The motivation for using and modelling with text is simple: many facts are not easily translated into summary numbers, and nuances are better expressed in text.

Steve’s objective was to propose a framework for using financial text and thinking about the sources of value associated with text. Clearly, even before starting, researchers should ask themselves about the comparative advantage of their research. Text mining packages such as those found in R, Python or SAS mean that the low hanging fruit has already been found, picked and eaten.

For example, when it comes to text analysis of 10-K and 10-Q reports for US firms, more than 60 papers have been produced. Once the added value of the project is clear, Steve proposes a four-step framework:

  1. Corpus Creation, corresponding to the Definition of the Problem
  2. Cleaning & Pre-Processing
  3. Annotation
  4. Processing, corresponding to the Search of Meaning.

For the Corpus Creation, the three broad genres of finance-related textual content are

  1. forum, blogs and wikis
  2. ii) news and research reports
  3. iii) content generated by firms.

Analysing multiple genres to contrast content views and triangulate results is likely to add more value. It is also important to decide whether to use the entire text or just part of it. Sharper conclusions may be possible by focusing on particular sections. Then one must harvest the textual data, clean it and pre-process it removing unwanted contents and organising the relevant unstructured text into structured text or numerical data arranged in tables.

The goal is to construct a term document matrix, i.e. TDM. Pre-processing may also include the removal of punctuation and numbers, removal of stop words, stemming, and disambiguation. Annotation may be manual or automatic and is critical for disambiguation and for feature extraction.

Manual tagging is more likely subjective but can play an important role for training in Big Text applications. In turn, automated tagging may use a number of available resources for Part-Of-Speech (POS), for morphology, grammar and syntax, for semantics and for pragmatic annotation.

Finally, processing may be based on simple word frequency counts using general dictionaries such as DICTION, General Inquirer, LIWC and others, or based on more refined approaches based on domain-specific lexicons such as Netlingo or Provalis.

There are many possible refinements in processing. Examples include handling categorisation, influence, emphasis words, specificity, similarity, obfuscation and fake news. Another axis often explored is weighting, according more importance to an unusual word or to words more closely related to the underlying construct.

The number of research papers on the value of text for investing keeps growing. Besides firm sentiment, some examples of successfully identifying firm- and manager-level risk factors include fraud and misreporting, CEO personality traits, idiosyncratic political risk, firm geographic exposure and financial constraints.

In Steven’s view, it is clear that automated text mining is a tool of increasing importance for quantitative investment.

For more articles by Raul Leote de Carvalho, click here >

 

Writen by Raul Leote de Carvalho. The post New data, new methods: highlights from the Inquire quant seminar – Part 1 appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

Despite concerns about slowing global and eurozone growth, low inflation and dovish central bank rhetoric mean investors’ search for yield, which has so far supported ‘peripheral’ eurozone government bonds, continues.

  • Spanish government bond market justifiably stable
  • Headwinds remain for Italian economy…
  • … and Italy’s political scene is not too stable either

Yield spreads between 10-year Italian and German government bonds widened to just below 300bp early in Q1 2019 after preliminary Q4 2018 GDP data (released at the end of January) confirmed that the Italian economy was in a technical recession. However, the 10-year Italian/German spread remained between 235bp and 275bp for the rest of the quarter. Spanish bonds were also well supported, with the spread between 10-year Spanish and German government bonds moving within a relatively narrow range of 95bp to 125bp over the quarter.

Spain – banking sector healthy, fiscal deficit progress

The stability in the Spanish government bond market is justified, in our view. Despite political fragmentation and minimal economic policy implementation over the last three years, growth in Spain has been largely unscathed by the slowdown in the rest of the eurozone. The banking sector is in good shape, with non-performing loan ratios falling back to 2010 levels. The fiscal deficit fell to 2.6% of GDP in 2018, and while this was a little short of the deficit target of 2.2% agreed with the European Commission, Spain had made significant progress in reducing the deficit to below 3% and achieved a close to flat primary balance.

More importantly, none of the political parties, including the nationalist Vox and far-left Podemos, is openly against the EU or the euro. This, and the progress on the fiscal front, makes Spain far less likely in the near term to get into confrontational situations with the EU about its budget. However, the Catalan issue is unresolved. And if the 28 April general elections fail to deliver a majority coalition government, the secessionist parties could gain more influence, halting fiscal consolidation and structural reform plans, and potentially putting Catalan separatism back on investor radars.

Italian economy – meaningful headwinds remain

In Italy, longer-term concerns about its fiscal health and political risks remain. While the European Commission decided against opening an Excessive Deficit Procedure against Italy in December, the Italian budget agreement remains far from ideal. The deficit targets are challenging given the country’s fiscal loosening plans and lack of economic growth. Aggressive fiscal relaxation in the form of universal income for the poor and the reversal of pension reforms were simply delayed and the funding of these programmes is unclear. The reduction in deficit targets for 2020 and 2021 relies heavily on “safeguard clauses”, in which a VAT increase is supposed to kick in automatically if no alternative fiscal measures can be found. The credibility of the current government’s fiscal discipline is low, and the Italian economy continues to face meaningful headwinds.

On the political front, since the coalition government of radical parties Five Star Movement (M5S) and the League was formed in mid-2018, support for M5S has fallen, while the popularity of the League has risen. Since the recent regional elections in central and southern Italy, the League is the dominant force within the centre-right coalition, garnering roughly 32% in the opinion polls versus 10% for Forza Italia.

Coalition ill at ease

National level differences in the M5S and League stances on these issues have led to a rise in tensions between the two governing parties:

  • the high-speed rail link between Turin and Lyon
  • devolution of fiscal powers from the centre to the north
  • the tactics on migration policy
  • joining China’s Belt and Road Initiative.

Given the League’s rise in popularity and its increased friction with M5S, there has been speculation that it can be only a matter of time before the League’s Matteo Salvini will want to break up the coalition, call fresh elections, and secure a clear mandate to pursue the party’s agenda.

In coming months, we see little appetite for a snap election. Such a move at this stage would mean asking the people to vote more or less at the same time as when the European Parliamentary elections take place. The two voting systems are quite different, requiring different coalitions, and the divergent expression of preferences would likely confuse voters and potentially taint the results of one or both elections. A summer election also looks unlikely as traditionally Italy has not held national elections after June.

Political risks will likely retake centre stage in the second half of 2019, particularly if Salvini does well in the European elections and sees it as evidence of his electoral advantage. In such a scenario, the League would likely become the senior partner in a government reshuffle or in a new centre-right coalition government. Market participants will likely perceive such an outcome as market-friendly, expecting that some of the expensive M5S fiscal agenda can then be dropped, and a new coalition would represent a return to political normality.

However, investors should remember that the anti-euro, anti-EU rhetoric over the last couple of years actually came from Salvini’s party. And regardless of who is in charge of the government, the budget discussion for 2020, which will also take place in the second half of this year, will again likely be challenging given Italy’s current fiscal health and the lack of progress on structural reforms.

Exhibit 1: Italian election polls

Source: Poll of Polls, April 2019

This is an extract from the Q1 2019 Inflation-Linked Bonds Outlook published in March. To read the full version, click here >

For more articles by Cedric Scholtes, click here >

For more articles by Jenny Yiu, click here >

 

Writen by Cedric Scholtes. The post Eurozone ‘peripheral’ markets: political risks remain appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

At the 10-year anniversary of the longest bull market in history for US equities, the S&P 500 Index has delivered 400% in total returns over the past decade. Many pension plans have benefited from rising equity markets and may now want to protect these gains.

At the recent Pensioen Pro Congress in the Netherlands, Koye Somefun, Head Of Multi Asset & Solutions in the Quant Research Group at BNP Paribas Asset Management, presented a strategy to protect equity gains with options. In this article, he explains why these strategies make sense for Dutch pension funds in the current environment.

It’s not just the perception of a high cost that makes pension funds hesitant to implement equity protection strategies with options. The timing of the deployment of put options and the loss of potential recovery power also cause doubts.

These misgivings became apparent at the congress during a workshop I gave on how pension funds can hedge their equity holdings against a stock market crash. [1] The main risk: equity portfolio losses

In my view, the greatest investment risk pension funds currently face concerns their equity portfolio.

In theory, interest rate risk is the greatest danger, but in practice, it is largely covered. For most pension funds, however, equity risk is completely unhedged.

Since the S&P 500 has now more than overcome the losses sustained during the Great Financial Crisis (see exhibit), this is not a problem in itself. Investors who opted for an equity protection strategy with options in recent years have actually achieved lower returns.

The question now is whether a pension fund can withstand large losses in the short term. In the event of a major sell-off, intervention by the central bank – as the industry supervisor – cannot be excluded.

The prudent option: protection

In my opinion, it may be prudent to put in place a temporary protection strategy with options. This can achieved with, for example, futures, put options or a combination of put and call options.

With an options strategy, the return on the S&P 500 from 2004 to 2018 was lower than without protection, but the underperformance during sell-offs was also smaller.

Currently, pension fund clients in the Netherlands do not use such protection strategies. This contrasts with insurance companies and pension funds in, for example, Germany and Italy, which actively hedge equity risk with options and/or futures.

At BNP Paribas Asset Management, we believe that option strategies are particularly suitable for Dutch pension funds because the regulator acknowledges they contribute to reducing equity risk.

Keeping the costs down and timing it right

During the workshop, it became clear that various Dutch pension fund managers have considered implementing a protection strategy for their equity portfolio. Ultimately, however, they decided against it for various reasons. It was not just the perceived high cost of put options.

Participants also noted that timing such strategies is difficult. “If we had bought put options in 2016 because of concerns over Brexit and the election of Trump, it would have cost us a lot of return”, remarked one manager. Rather than falling, stock markets have risen further since then.

Adam Barszczowski, board member at the pension fund for physiotherapists, pointed out that an equity protection strategy leads to lower return expectations for equities. This is a difficult subject for his pension fund because it has a funding ratio of around 100%. “We need the full upside potential of equities to be able to maintain the fund’s recovery power.”

Pensioenfonds PGB has said that it considers an equity protection strategy to be compatible with its role as a long-term investor.

Appealing options to contain the costs

In the current environment, I believe protecting equity gains with options is well worth considering. Funds with a funding ratio of more than 105% will want to prevent it from falling below that level. A protection barrier for equities can help with this. Additionally, a pension fund would not necessarily need to finance all of this protection by giving away some equity upside.

A well-constructed protection strategy can significantly reduce the costs. Additionally, pension funds could look inside their multi-asset portfolios at other trade-offs. For example, partly financing the equity protecting by (asymmetrically) increasing their interest rate hedge via swaptions (an option on an interest rate swap) could be an interesting alternative. It means less upside in the case that interest rates rise, but more exposure to rising equity markets.

[1] ‘Winter is coming’. How do you prepare your portfolio for winter?
Spring is almost here, but according to many analysts, the prospects are particularly harsh. Some are quoting ‘Game of Thrones’: ‘Winter is coming’. How can managers best prepare their pension funds?; Pensioen Pro Congress workshop on 14 March 2019

For more articles on equities, click here >

 

Writen by Koye Somefun. The post Protecting equity gains with options appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

The tone of the March ECB meeting was more dovish than anticipated, with three key pieces of new information grabbing the markets’ attention.

  • Change in forward guidance – further interest rate rises before 2020 explicitly ruled out
  • A new Targeted Longer-Term Refinancing Operation (TLTRO-III) to support bank lending conditions
  • Larger-than-expected downgrades to the ECB’s growth and inflation projections

Although the above announcements were unexpected either in terms of scale or timing, it is unlikely that they will meaningfully ease financial conditions in the eurozone. Why?

First, the change in the forward guidance on interest rates – from keeping rates unchanged “at least through the summer of 2019” to “at least through the end of 2019” – was modest. Market pricing had already pushed back expectations for the first rate rise into 2020 ahead of the meeting. In fact, several council members presented the option of changing the date of the forward guidance to March 2020, but this was resisted due to concerns over the consequences of a protracted period of negative rates and the associated implications for banks.

Second, the news on TLTRO-III lacked detail. The new series will be launched in September and expire in March 2021, each with a maturity of two years. The interest rate will be indexed to the rate on the main refinancing operations over the life of each operation. ECB President Mario Draghi announced there will be built-in incentives for credit conditions to remain favourable, but did not give details. With the last tranche to be originated as late as in March 2021, this ensures that bank lending conditions will remain favourable until March 2022, when the residual maturity of these loans falls to below one year. However, compared with previous TLTRO incarnations of three to four-year fixed rate loans, TLTRO-III is less generous and sends a weaker signal about the future path of policy rates.

Lastly, ECB staff economic projections were markedly less optimistic than those in December 2018. Calendar year growth for 2019 was revised down from 1.7% to 1.1%. Looking at the quarter by quarter projections, it is not surprising to assume weak growth in the first half of 2019 given recent data releases. Real GDP growth is projected to remain low in the second half, at around 0.25% quarter-on-quarter (QoQ) in both Q3 and Q4, roughly half December’s estimates.

While the downward revisions could send a dovish message, signalling more easing in the pipeline, the correct interpretation could be more nuanced, particularly as the ECB took action alongside weak forecasts. The latest projections establish a new economic baseline against which data outturns can be evaluated, and policymakers may argue that they need to see weakness relative to that new baseline before being convinced to do more.

The fact that further cuts in interest rates to deeper negative levels, or a resumption of quantitative easing, were not even discussed at the meeting suggests that the council has a relatively sanguine view about the current state of the economy. This contrasts sharply with the US Federal Reserve’s pivot over the past quarter.

The path of inflation returning to target – delayed, not derailed?

President Draghi’s description of the outlook remains one where growth weakness and downside risks arise mainly externally, with limited signs of spill-overs into domestic demand so far, and sustained convergence of inflation to the target has been delayed rather than derailed.

The inflation-lined bond markets seem to think differently. Having spent much of 2018 gyrating between 1.6% and 1.8%, the 5-year/5-year forward breakeven inflation (BEI) rate started to fall precipitously in Q4 2018 and reached a low of around 1.3% in March 2019, just several basis points above its all-time low in mid-2016. The decline is large by historical standards, and reflects both the expectations that eurozone inflation will remain low and scepticism of the ECB’s credibility on inflation. So far, Draghi has dismissed talk of inflation expectations being unanchored by referring to the stable survey-based measures, while pointing to a negative inflation risk premium for driving market-based expectations.

Exhibit 1: Market-based vs. analysts’ inflation expectations in five years’ time

Source: Bloomberg, March 2019

It is reasonable for investors to worry about the ECB being late to ease and/or running out of ammunition. The political barriers to the ECB returning to unconventional monetary easing are high, and reaching consensus on the need for stimulus and then calibrating it is time consuming. We know the ECB can ease more if needs be, but how much pain might it take to force the ECB into action, particularly when the leadership is about to change this autumn?

Signs of the ECB preparing to do more, if necessary

Fortunately, there are signs that the ECB is preparing to do more if weak external factors spill over more significantly into domestic demand. First, the council retained the downside skew on the growth outlook, which could be understood as a signal that it is not done easing.

More recently, the discussion on deposit rate tiering restarted. Draghi hinted in late March that the ECB will reflect on possible measures to preserve the favourable implications of negative rates for the economy while mitigating the side effect of hitting banks’ profitability. One ECB board member later acknowledged that a “lower-for-longer” environment had triggered the discussion of a tiering system and “ECB sources” also revealed that technical committees are analysing options for such a system.

While it is too early to suggest that the ECB will be shifting its policy stance significantly, an adoption of a tiering mechanism would send a dovish signal. Tiering would probably be adopted alongside a “lower-for-longer” policy, and would also lend credibility to threats of further rate cuts.

This is an extract from the Q1 2019 Inflation-Linked Bonds Outlook published in March. To read the full version, click here >

For more articles by Cedric Scholtes, click here >

For more articles by Jenny Yiu, click here >

For more articles on the ECB, click here >

 

Writen by Cedric Scholtes. The post The ECB turned more dovish in March appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

The slump in eurozone new manufacturing and export orders is countered by service sector reslience and healthy domestic demand on the back of wage growth and fiscal stimulus.

  • Hard manufacturing data less pessimistic than business surveys
  • Rises in employment, higher pay and lower energy costs imply consumption support
  • Impact of recent external shocks should fade

Macroeconomic development in the eurozone has remained disappointing, with slowing growth extending into 2019. Support for GDP growth from abroad remains a headwind: slowing global manufacturing activity in an environment of high and rising political and policy uncertainties bodes poorly for manufacturing in the eurozone.

Indeed, the March survey of purchasing managers pointed to more bad news in the manufacturing sector. The PMI index dipped from 50.5 in January to contractionary levels of 49.3 and 47.6 in February and March, respectively. Turning to the forward-looking aspect of the data, we find new orders unchanged on the month at the symbolic 50.0 “no change” point. But the sectoral split shows that sector weakness is evident – the indices for both new manufacturing and export orders slumped to below 45.

Hard data, although lagging, painted a less pessimistic picture. January’s industrial production grew by 1.4% month-on-month (MoM), which is 0.4% above the average for the previous quarter.

The indicator for new passenger car registrations posted its fourth consecutive increase in January, confirming expectations for a normalisation in car registrations after the disruption caused by the implementation of new car emission standards. Although there are undeniable signs of external risks to growth in the manufacturing sector, it remains to be seen whether the hard data will reflect the same scale of contraction in activity as implied by the PMI surveys.

Consumer confidence improves…

On a positive note, the eurozone’s services sector is showing signs of resilience. The services PMI bounced from a low of 51.2 in January to 52.8 and 52.7 in February and March, respectively. The new business balance for services improved to 52.1 in March – its highest since November 2018.

The latest economic indicators are pointing to steady growth in private consumption. Most eurozone countries saw broad-based increases in employment in Q4 2018. The cumulative growth in employment, coupled with rising wages and the recent drop in energy prices, implies steady growth in households’ real disposable income.This should continue to support consumer spending.

Indeed, consumer confidence rose for a second consecutive month in February, halting the decline in most of 2018. This is consistent with ongoing steady growth in private consumption.

…but improvements in external factors are also needed

While the resilience of domestic growth is comforting and will likely act as an economic shock-absorber in the near term, a deeper-than-expected slowdown could eventually cloud the outlook for the labour market and cause consumption to slow.

An improvement in external factors will thus be needed to turn the economic momentum from its current slowing trend. We see an increasing likelihood that the uncertainty shocks which have hit the global economy and markets in recent months should fade. First, stimulus efforts by the Chinese authorities should help Chinese growth to bottom out. Second, the trade truce and an emerging trade deal between the US and China should help to halt the slowdown and disruption in global trade, although the timing and magnitude of a deal remains uncertain.

Third, while the Brexit situation remains fluid, the removal of an immediate threat of a disorderly Brexit should help to reduce the level of economic uncertainty in the eurozone in the near term.

Eurozone purchasing managers’ index (PMI) fell further in Q1 2019

Source: Haver, as of March 2019

Forward-looking PMI – new orders

Source: Haver, as of March 2019

Hard data vs. soft data – manufacturing sector

Source: Haver, as of March 2019

Balance remains above its historical average

Source: Bloomberg, March 2019

For more articles by Cedric Scholtes, click here >

For more articles by Jenny Yiu, click here >

For more articles on the eurozone, click here >

 

Writen by Cedric Scholtes. The post Sluggish eurozone growth continues into 2019 appeared first on Investors' Corner - The official blog of BNP Paribas Asset Management.

Read Full Article

Read for later

Articles marked as Favorite are saved for later viewing.
close
  • Show original
  • .
  • Share
  • .
  • Favorite
  • .
  • Email
  • .
  • Add Tags 

Separate tags by commas
To access this feature, please upgrade your account.
Start your free month
Free Preview