You are on page 1of 11

October 2012

Strictly Private and Condential

Investor Letter

Prot Margins A Broader Perspective


Alan Burke Co-Chief Executive

Summary
The French mathematician Benoit Mandelbrot (1924-2010) is most famous for introducing the world to fractals shapes that look the same under any magnication. You can zoom in on a fractal a thousand times and it will retain the same pattern and detail. You can get completely lost in them and their timeless beauty (log into http://mathworld.wolfram.com/Fractal.html for additional details). It turns out fractals occur everywhere in nature, from leaves to lightning and from snowakes to seaweed, but it took Mandelbrot to point them out. Sometimes we become so accustomed to looking at something that we fail to see its true patterns. On the other hand, sometimes it helps to change your scale and perspective to see the true pattern emerging. Unlike fractals, Prot Margins are a phenomenon where an appropriate focus matters. At the right scale (i.e. long term in scale and broad in its scope), trends become apparent that may not necessarily be so obvious when caught up in recent noise. The chart on the right shows Avocas GPM indicator. GPM is our measure of Global Prot Margins, a 50 year dataset which we have built up using prots data relative to GDP from national accounts as well as exchange rate and price data taken from an array of national and international agencies. GPM has clearly had a strong run in recent decades. So what does this mean? In nature, gravity limits the height of trees, and likewise in the nancial world there are numerous forces, cyclical and secular,

Avoca Global Prot Margins Indicator (1960=100)


140 130 120 110 100 90 80 70 60 50 40
1970 1960 1980 1990 2000 1965 1985 1995 2005 2010 1975

Source: AMECO, OECD, and national statistical agencies. The chart shows the trend in aggregated national operating surplus as a percentage of GDP

limiting the growth of Prot Margins. In everyday life, we call this phenomenon Mean Reversion. In Avoca, Prot Margins that are substantially above the long term mean make us nervous. As we shall see, Mean Reversion will inevitably reduce Prot Margins in the not too distant future. Reversion to the Mean of Prot Margins will have implications at the company, sector, country and economy wide levels. This will inevitably have implications for Avoca and other investors in terms of individual name selection, portfolio construction and risk management.
www.avocacapital.ie

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

October 2012

Prot Margins - A Broader Perspective


Introduction
In the immediate aftermath of Lehman, on being told that the EBITDA performance of our portfolio companies was holding up better than might be expected, the most frequent comments made by Avoca investors were Great, but how is that happening? and Is it sustainable? In response to those questions, at the time, we did additional work for our investors on the 150 names in the Avoca portfolio which showed amongst other things that: The average EBITDA margin for the portfolio was 23%, with the remaining costs above EBITDA representing 77% of Revenue split roughly equally between Labour and Raw Materials. In the year post Lehman, Avoca portfolio Revenues declined by 4-5% (roughly in line with European GDP). EBITDA margins for the portfolio were maintained (which meant that absolute portfolio EBITDA also broadly declined in line with GDP). EBITDA margin maintenance for the Avoca portfolio was due to roughly equal amounts of savings in both the Labour and Raw Material cost categories.

Understanding Prot Margins - The Classical Approach


When I was in university, over 20 years ago now, Economics 101 referred to three factors of production; Land (a category including both Energy and Raw Materials as well as Rent), Labour and Capital. The basic concept was that these three factors are the elemental forces that in combination are needed to drive economic output and activity. More recently, Technology is sometimes quoted as a fourth factor of production. Given the disparity between the retail price of an Apple smartphone at approximately $800 and its total cost of materials and labour of substantially less than $200, one can see why technology might be deemed an integral factor of production. However, I still prefer the old three pronged classication as technology can be incorporated into the old framework as being in either the Labour or Capital camp depending on ones point of view. At a very basic and accounting level, Prots are easy to describe. They are simply Revenues less Costs. However, from an economic theory perspective they are something much harder to dene. The exact nature of what Prots are in a society has concerned economists ranging from Ricardo to Marx, going as it does to the very core and essence of Economics. It is generally accepted today that prots represent the (necessary) return on the capital that has been employed in an economy without which it would not be possible to induce the owners of that capital to accumulate it and put it to risk in the rst place. Using the very simple 3 basic factors of production model as a framework, it can be shown that there are a number of factors that aect the level of Prots earned in an economy including most importantly: The degree to which a society is capitalist or socialist in nature; The relative supply and demand dynamics of Land, Capital and Labour; and The Economic cycle.

One nagging doubt remaining from the work we did at that time on the Avoca portfolio was the question of the sustainability of Prot Margins at the economy wide level. If one of the principal sources of Prot Margin maintenance for our portfolio was Labour cost savings then surely for the economy as a whole this becomes unsustainable as one persons cost saving is anothers loss of income. Or put another way, how could it be possible for the economy as a whole to cost cut its way to higher levels of protability? Three years on, with Prot Margins perceived to be at or close to all time highs, we have decided to re-visit the entire question of Prot Margins with a more in-depth look at the issues including hopefully answering important questions such as: What are the factors that drive Profit Margins for the economy as a whole? Where do Profit Margins sit today relative to history? and How likely is it that Profit Margins might decline in the future and what might the catalysts for any such decline be?

In attempting to answer these questions, we will take a broadly top down approach to the problem focusing rstly on the Theory of Prot Margins before examining the evidence from the available data as to what has actually occurred. Finally, we attempt to draw some useful conclusions from our research, focusing on its potential implications for macro views, and Avoca portfolio construction and risk management.

There are undoubtedly other factors that will also aect Prot Margins such as; productivity (e.g. imagine the eect on global Prot Margins if a low cost reliable source of energy such as ultra-ecient solar power could be invented); the mix of sectors in the economy (e.g. an economy dominated by naturally high margin sectors such as pharmaceuticals, software or energy will have higher Prot Margins all else being equal) and the degree of competitive intensity (monopolies will naturally make for higher Prot Margins).

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our 2 authorisation and regulation by the Financial Services Authority are available from us on request

October 2012

Prot Margins - A Broader Perspective


a) The degree to which a society is capitalist or socialist in nature Prots are what accrue to Capital after its economic battle with Land and Labour has been fought (as refereed by government). Dened as such, we see that the nature of society (i.e. more socialist or capitalist), will be one of the key determinants of prot levels. For example, at the extreme, in a communist society, returns to Land and Capital should be nil as both have been expropriated and all the spoils of economic activity accrue to Labour. In preparing this research Avoca looked at Prot Margins from the National Accounts data of over 40 countries stretching back over 50 years. The results for 2011 Prot Margins by country are shown in the chart below. Broad Prot Margins
80% 70% 60% 50% 40% 30% 20% 10%
Iceland Sweden Denmark France Israel Russian Federation United Kingdom United States Switzerland Canada Slovenia Japan Belgium Portugal Netherlands Finland EU Germany Austria EZ Australia Estonia New Zealand Spain Hungary Korea South Africa Italy Ireland Norway Czech Republic Poland Slovak Republic Chile Greece Mexico India

US Eective Corporate Tax Rate


45 40 35 30 25 20 15 10 5 0

1972

1986

1980

1996

1990

1984

1988

1994

1998

1976

2006

2000

1974

1978

2004

2008

1992

Source: Congressional Budget Oce and Bureau of Economic Analysis

0%

Source: OECD and national statistics agencies (broad margins show prots before depreciation, taxes and some other expenses these are not subtracted because of inconsistencies between countries in how they are handled broad margins are comparable across countries)

The level of tax imposed by a government on prots will obviously signicantly impact Prot Margins. Higher levels of globalisation, the relaxation of capital controls and greater tax competition have all undoubtedly contributed to Capitals taking a greater share of the economic pie over the past 30 years or so. Greater levels of global capitalism have certainly benetted owners of Capital. b) The relative supply and demand dynamics of Land, Labour and Capital Positive or negative shocks which aect either the supply of or demand for Land or Labour should impact the returns to Capital. As an example, one would expect that were a severe oil shock such as happened in 1973 to re-occur there would be an inevitable reduction in global prot levels as economic resources would be redistributed to oil producing countries. The graph below shows the impact of recent oil price moves on US Prot Margins. The three major spikes in oil prices, beginning in 1973, 1978 and 2007, all coincide with downward falls in Prot Margins. US after tax Prot Margins versus Nominal Oil Prices
160 140

The results threw up some thought provoking anomalies, like the generally high margins in Portugal, Ireland, Italy, Greece and Spain (we are not claiming any cause and eect!). Other countries Prot Margins were less surprising, such as low margins in Japan and high margins in oil or mineral exporters like Norway and Chile. The importance of a countrys political system is clearly seen in the chart. The really high Prot Margin countries are those with a high degree of income inequality due to political factors, including some former Eastern bloc countries, Mexico and India. In addition, one thing that will also strongly support Prot Margins is a monopoly, and Mexico for example has quite a few of those. This is a case where the government referee may be biased towards Capital. The impact of taxation Closely related to the capitalist or socialist nature of society is the level of taxes that are imposed on capital by government. The following graph for example shows the eective tax rate on Corporate Prots in the US over the past forty years.

2002

2010
12 10 8 4 2 0
Jul 11 %

1982

2007 price spike (speculation, geopolitics, etc)

120

100

1979 Iranian oil crisis

US$

80

60

40

20

OPEC oil embargo, 1973

0
Jul 71 Apr 70 Jan 79 Jan 74 Jul 86 Jul 96 Jul 06 Jul 76 Oct 82 Oct 92 Oct 02 Oct 72 Apr 75 Oct 87 Oct 97 Apr 80 Oct 07 Oct 77 Jan 69 Jan 89 Jan 99 Apr 90 Apr 00 Apr 85 Apr 95 Apr 05 Jan 09 Jan 84 Jan 94 Jan 04 Apr 10 Jul 81 Jul 91 Jul 01

Oil Price (left axis)

After tax Prot to % GDP (right axis)

Source: U.S. Department of Commerce and Bureau of Economic Analysis, West Texas Crude Oil price from Dow Jones and Federal Reserve Bank of St Louis

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request 3

October 2012

Prot Margins - A Broader Perspective


These oil shocks were upward input cost shocks and downward demand shocks a double hit for Prot Margins. Globalisation can be viewed as a downward input cost shock, and should have had the eect of increasing Prot Margins by increasing the global supply of lower cost labour. As we shall see in detail later, globalisation has been a boon for certain sectors that have been able to use the availability of emerging market labour to reposition their cost base. c) Economic cycles Margins are very closely linked to the business cycle. This pattern is very regular as evidenced by the chart below which shows the evolution of US after tax Prot Margins since the 2nd world war (the shaded areas represent recessions). US after tax Prot Margins
12%

Another way of looking at Economy Wide Prot Margins The Kalecki equation
Michael Kalecki (1899-1970) was a Polish born economist who at various stages in his career held economic tenures at Cambridge and Oxford before settling in the later stages of his career as a professor in Warsaw in what was then communist Poland. Today he is better known amongst the economist community for the Kalecki equation, a way of looking at economic data that has proved to have useful insights for studying economy wide Prot Margins. Kaleckis insight was to view business prots in terms of the net spending (or saving) of businesses, households, government, and the foreign sector. The full derivation of the Kalecki equation is included in Appendix 1 and can be summarised as follows: Prots = Investment +Dividends H(s) -G(s) -F(s) Where H(s) is the household saving rate, G(s) is the government savings rate and F(s) is the foreign savings rate (better known as the current account decit). The Kalecki equation inputs for the US for 2011 were as follows:

10%

8%

6%

Input Investment Dividends Household savings Government Savings


1959 1955 1971 1999 1951 1979 1963 1983 1995 2003 1975 1947 1967 1987 1991 2007 2011

% of GDP +3.2% +5.5% +3.2% -7.6% +3.1% (US current account decit)

4%

2%

0%

Foreign Savings
Source: US NIPA, Flow of Funds 2011

Recession

US after tax Prot Margins

Data run to Q2 2012. On current estimates, margins are down to 9.5% from their 10.3% peak in Q4 2011. Source: U.S. Department of Commerce, Bureau of Economic Analysis, and Federal Reserve Bank of St Louis

Like most strong patterns, there is a good reason why Prot Margins and the cycle are interlinked. As a recovery progresses, demand for Land, Labour and Capital rise, causing the costs of these inputs to rise. This eats into Prot Margins before the economy actually starts to slow. In fact, Prot Margins tend to peak approximately half way through the cyclical upswing. Given that we are now three years into the current economic cycle, and the average period between recessions since WWII has been six years (including 2 unusually long recession free periods between 1982-1990 and 1991-2001), there are clear reasons to believe that Prot Margins may have reached a cyclical peak for the present cycle.

Plugging these inputs into the Kalecki equation, US after tax Prot Margins for 2011 as a percentage of GDP were: Prots= 3.2% + 5.5% - (3.2%) (-7.6%) (3.1%) = 10.0% GDP Dividends and investment both sent money back into the economy, supporting prots. After their credit boom debt binge, households in 2011 decided to pay down debt (i.e. net savings) to the tune of 3.2% of GDP a negative for prots all else being equal. The foreign sector was also a net drag on US prots in 2011. Since the US is a net importer of goods and services, the prots on such goods and services accrued to overseas companies rather than US ones. The big driver of Prot Margins in 2011 obviously was the government sector, which bankrolled prots to the tune of 7.6% of GDP. We started by asking if Kalecki could tell us why Prot Margins have had a relatively good time through the Great Recession. The answer is that public sector largesse has undoubtedly supported Corporate Prot levels. Here it gets potentially scary, because we know that scal consolidation is the current game plan for Europe and is also potentially looming as we know on Capitol Hill.

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request 4

October 2012

Prot Margins - A Broader Perspective


The importance of monetary policy on the Kalecki framework.
At the pleadings of our in-house economist Aidan Corcoran, we include one other important factor aecting Prot Margins - monetary policy and global interest rates. Aidan argues not unreasonably that monetary policy has been a big factor in keeping the global economy out of a 1930s style depression, which clearly would have had a strong eect on Prot Margins. In support of his argument he makes the point that QE is currently funding the major part of the US and UK scal decits. By September 2012, QE had provided a home for 375bn of mostly government bond assets in the UK and $2.3tn of similar assets in the US (representing around 100% and 60% of the last 3 years scal decits in the UK and US respectively). Kalecki tells us a big factor supporting Prot Margins is Fiscal decits, but these decits are currently very strongly supported by policy action from central banks. Fiscal decits are not doing all of the work to support Prot Margins on their own. can be heavily distorted by periodic episodes that can lie far from the mean. This insight means that we have to be inherently careful in assessing whether Reversion to the Mean will happen as of right (and crucially when it will happen) for any economic dataset including Prot Margins. One of the implications in measuring any phenomena that might behave in a Mandelbrot way or with fat tails is that the length of the data set and the starting point matter hugely. The chart below shows Broad Prot Margins since the US Civil War (courtesy of Dumenil and Levy - the Prot Margin shown is not comparable to our GPM because of data dierences, but the trends are). Broad Prot Margins since the US Civil War
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%

A nal thought - Mandelbrot and the potential implications for Prot Margins
Basic theory (supported by the available evidence) highlights a number of drivers of economy wide Prot Margins. Before concluding as to where Prot Margins go from here, it is useful to visit Mandelbrot again whose musings about the nature of nancial markets have relevance for the question of examining the behaviour of Prot Margins. Joseph and Noah When he was not busy writing about nature in equations, Mandelbrot found time to study nancial markets in detail. He described the nancial world as being dominated by two broad eects: the Joseph eect and the Noah eect. The Joseph eect refers to the common biblical allusion of 7 years of feast followed by 7 years of famine. This is also better known as Reversion to the Mean. The Noah eect is slightly less intuitive. It refers to the fact that sometimes as in the bible it does rain for 40 days at a time (here in Ireland we didnt need the bible to learn that). Another way of describing this is that sometimes apparently unusual events happen and when they do, they have a tendency to cluster together. Examples of these Noah like phenomena that have been documented in nature include oods, hurricanes and earthquakes. The underlying phenomena at work causing these cluster eects are usually feedback loops of some degree. In the nancial world it shows up in stock market returns that are not normally distributed otherwise known to us as fat tails. The relevance to our discussion is that while under economic theory Prot Margins should be intuitively mean reverting, the mean itself

1889

1884

1939

1934

1949

1959

1954

1869

1899

1969

1989

1944

1894

1984

1909

1999

1964

1904

1994

1879

1874

1979

Source: Dumenil G. and Levy D., The U.S. Economy since the Civil War (1994, data updated in August 2010)

The early spike coincides with the post-Civil War era, when the rule of law was fragile (prompting President Lincoln to install military governors in many states). Lawlessness and corruption tend to enrich the few in power (Capital) instead of enriching the many (Labour). Clearly war has, on at least three occasions in the past, supported Prot Margins in the US and the Great Depression was a cardiac arrest for them. Neither wars nor depressions could have necessarily been foreseen with any certainty, but both happened, and both impacted Prot Margins hugely (popularly known today as the Black Swan eect). The implications of the above are philosophical and practical in nature for us. Prot Margins, like many nancial markets are not represented by classic bell curve distributions. They can be subject to long periods of relative calm and low volatility with the potential for extreme outcomes on occasion. The challenge to us as investors is to remain cognisant of the potential for underlying volatility and its potential impact on our Margin of Safety in making any investment decision.
www.avocacapital.ie

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

2009

1974

2004

1919

1914

1929

1924

October 2012

Prot Margins - A Broader Perspective


The Evidence
1) Prot Margins across the globe are close to 50 year highs Our methodology in assessing where global Prot Margins sit relative to their historical averages was to focus on National Accounts data rather than market data (e.g. S&P 500 or FTSE etc.). The reason for this was to allow ourselves as large a data set as possible on a broadly comparable basis which avoids distorting issues such as diering accounting policies, numerator and denominator eects (e.g. comparing the Prots on the S&P 500 with US GDP for example) and tax driven measurement issues. Our sample size was over forty countries currently accounting for 70% of global GDP. The path of Global Prot Margins (GPM) for our dataset since 1960 is shown below. Avoca Global Prot Margins Indicator (1960=100)
140 130 120 110 100 90 80 70 60 50 40

Prot Margins in the future? Our sectoral analysis below suggests that when analysed on a sector by sector basis, the answer is most probably no over the medium term - certain sectors have benetted greatly from recent trends such as globalisation and the inevitable forces of capitalism will draw more resources towards Labour and away from Capital over time. The above graph together with the previous evidence for long term US Prot Margins also highlights how vulnerable global Prot Margins are to potential negative supply shocks- the 1970s oil shock for example caused Global Prot Margins to decline by a quarter- a salutary thought given the prospects of further tensions in the Middle East in 2013 (even if the global economy is less dependent on oil now versus then). 2) Certain sectors have benetted disproportionately from globalisation National accounts data helped us assess where global Prot Margins are relative to their historical perspective which we believe is an extremely valuable insight. However, for our investment implications we focussed more on market data, including rms in the S&P 500, FTSE etc. Here we analysed Prot Margins for over 2000 companies in 20 sectors (and numerous sub-sectors) over a 30 year period. One factor we were interested in quantifying was the impact of globalization on Prot Margins. We believe certain sectors, which we have termed the globetrotters, have globalized particularly aggressively over the last twenty years. The Globetrotters shown in the light blue bars include Capital Goods, Semi-conductors, Retail, Software and Technology equipment sectors. It is no coincidence that our analysis shows all these sectors to have grown Prot Margins, with two of them among the top three gainers. We have highlighted these sectors in the chart. EBITDA Margins in 2011 vs 30 year averages with changes
Average Margin versus Latest Margins
Energy Software & Services Semiconductors & Semiconductor Food Beverage & Tobacco Pharmaceuticals, Biotechnology Materials Technology Hardware & Equip. Retailing Health Care Equipment & Service Media Household & Personal Products Consumer Services Capital Goods Globetrotters Other Sectors

1970

1980

1960

1990

2000

1985

1965

1995

Source: AMECO, OECD, and national statistical agencies. The chart shows the trend in aggregated national operating surplus as a percentage of GDP

Prot Margins, despite having experienced a cyclical downturn following the Great Recession, remain well above their recent historical norms. In fact given the extent of the global downturn since 2007, one would have expected that Prot Margins might have gone through the oor over the past 5 years. The fact that they have not suggests that there must be very strong secular forces at work as well as the extreme scal and monetary policies supporting the global economy post 2008. Secular factors that have caused Prot Margins to behave in this manner over the past 50 years, and in particular to have maintained Prot Margins at elevated levels over the past 5 years, are myriad but undoubtedly include globalisation, the increasingly capitalist nature of society, lower global corporate tax rates, declining global interest rates and increased productivity. The question for us is will these secular factors continue to support

2005

2010

1975

Change in Latest versus Average

Commercial & Professional Service Transportation Automobiles & Components Food & Staples Retailing Telecom Consumer Durables & Apparel Utilities 0 5 10 15 20 25 30 35 -2 0 2 4 6

Average (Bars)

Latest (Dots)

Source: Bloomberg, Avoca database. published company accounts

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request 6

October 2012

Prot Margins - A Broader Perspective


Across the twenty sectors, average 2011 EBITDA Margins increased from 18.1% to 19.9%, a c.1.7% increase over the long term average or in absolute terms c.9.4% growth. Given the impact of lower global interest rates and corporate taxes as discussed above, the increase in after tax Prot Margins over the long term average was most likely substantially in excess of 9.4%. The globetrotters increased their EBITDA Margins to an even greater extent, from 14.9% to 18.7%, a 3.9% gain or 26% growth in the absolute Prot Margin level. This most likely derives from greater use of low cost labour through o-shoring and outsourcing over the past thirty years. The substantial increase in EBITDA margins across the entire 20 sectors has surprisingly not been accompanied by an increase in the capital intensity of those sectors - in fact the opposite has occurred. Overall, the Capex/Sales ratio of the above 20 sectors decreased from a long term average of 7.5% to 6.1% in 2011, a 1.4% decrease. Combined with the 1.7% increase in average EBITDA margins above, Free Cashow to Sales has increased by 3.2% from 10.6% to 13.8% over the period- a substantial 30% increase in Free Cashow. Another way of putting this, is that the Return on Capital across the entire 20 sectors must have substantially increased over the last 30 years. It has indeed been a good time to be a capitalist. Globalisation has clearly been a boon for certain industries that have been able to reposition their cost base. However, classical theory would suggest that such Prot Margin gains are likely to ultimately have to be shared by Capital with Labour. The question is over what time frame? Recent labour unrest in China (Foxconn) and South Africa (Lonmin and Anglo Gold) may be indicative of this phenomenon at work in real time. The Lonmin Labour factor could quite possibly accelerate in 2013 driven by an increase in global food prices (up 8% YTD in dollar terms). Prot Margins in sectors exposed to the Lonmin factor are potentially at risk in the short to medium term. 3) Returns to Land have also increased substantially in recent years We also looked in detail at the Energy and Materials sectors to get a proxy for the returns to Land, completing our analysis of three factors of production model. The last 30 years have also been good for Land and natural resources. Average EBITDA margins for the energy and materials sectors increased from 20.9% to 24.4% over the period of our study, a 17% increase in absolute terms. These two Sectors should in theory be subject to the iron clad law of Reversion to the Mean given that they are the classic denition of commodity sectors. Any increase in protability should have been
75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129 25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399 www.avocacapital.ie

accompanied by an increase in the capital intensity of the sector or it should over time have been competed away by new entrants and new supply to meet the extra demand. This did happen to an extent for Energy, a 4.9% increase in EBITDA margin was accompanied by an increase of 0.8% in the Capex/Sales ratio. Exploration cost increases did partially oset increased margins although not to the degree one might expect. However for Materials, an increase in margins from 24.1% to 26.4% was accompanied by a decrease in Capital expenditure intensity from 11.5% to 11.3% of Sales. Taken together, the Free cashow for these 2 sectors increased from 9.9% of sales to 13.3%, a 33% increase. The owners of Land have also done well over the past 30 years. 4) Getting more granular In order to help develop practical implications for our research on Prot Margins, we have drilled down further into Prot Margin data on a sector and sub-sector basis in order to examine the trends in individual companies Prot Margins over the period of our study. By way of illustration, our graphs below show the evolution of EBITDA margins and Return on Invested Capital (ROIC) over the past 25 years for the 25 largest companies in the Diversied Metals and Mining sub-sector. Each dot on the graph represents the results from a particular company in that sector for that particular year. EBITDA Margins for Materials Sector
40 35 30 25 20 15 10 5 0 -5
1991 1997 1988 1994 2000 2006 2009 1985 2003 2012

Big

Small

Note: Materials Sector = Diversied Metals and Mining Sector. Source: Avoca database, published company accounts and Bloomberg. Green indicates the largest 50% of companies by revenues.

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

October 2012

Prot Margins - A Broader Perspective


ROIC for Materials Sector
60 50 40 30 20 10 0

Conclusions
Our approach to looking at Prot Margins involved two broad approaches, aggregation of country and time series data to identify the bigger macro picture and a disaggregation of data at the sector and company level to allow us to identify trends that might provide useful insight for portfolio construction and risk management. Our conclusions are as follows: General insights 1. Unlike fractals, understanding the nature of Prot Margins is greatly inuenced by perspective. A broader, and in particular a more long term perspective, is fundamental for understanding the nature and drivers of Prot Margins. 2. Prot Margins are much more complex than meets the eye. When analysing Prot Margins, trends, start and end dates matter as does sample size, sector and country composition. Most research we have seen on Prot Margins has been conned to country or sector level analysis. Avoca needed to build its own index to understand the global trend which is shown for the rst time in this newsletter. There are large variations in trends in Prot Margins at country and sector level, and these trends can be quite long term in nature. This hinders attempting to directly prot from Mean Reversion because for very long periods there may be no obvious mean. Prots do not necessarily behave in a sine wave like manner.

1988

1994

2006

Big

Small

Note: Materials Sector = Diversied Metals and Mining Sector. Source: Avoca database, published company accounts and Bloomberg. Outlying high values tend to be for rms concentrating on copper mining.

2009

2000

1985

2003

1997

2012

1991

It can be clearly seen that notwithstanding a decrease post the Great Recession, Prot Margins and ROIC in the diversied metals and mining sector remain elevated today. There are broadly three schools of thought on the above: The above rise in Prot Margins and ROIC is as a result of a nite supply of Materials meeting exponentially increasing demand from China and other developing regions scarcity and demand has caused Prot Margins and Return on Invested Capital to go to a permanently higher plateau (the super cycle theory). After a long period of supply lagging demand, ongoing capital expenditure will result in supply catching up with demand bringing ROIC for Materials down to more normal levels. (Anecdotal evidence would suggest that despite the current downturn there is still substantial ongoing late-cycle capital expenditure in the Materials sector in economies such as Australia, South Africa and Chile). Materials have benetted from an investment bubble in China in particular and the above excess Prot Margins and ROIC are the harbinger of a coming bust in the Materials sector.

3.

4.

Macro insights 1. Margins are high by historical standards and face a number of short term headwinds. These headwinds include: 2. An overdue cyclical correction which has consistently occurred in Prot Margins mid cycle following a recession; An extremely high sensitivity to Government policy (scal and monetary); and Traditional sources of cheap Labour appear to be starting to demand a greater share of the economic pie.

Two of the above three points (i.e. the second and third bullet points) would point to a mild to signicant slowdown in Prot Margins and ROIC for the Materials sector in the short term at least. Such concerns regarding the sector are possibly among the reasons why the Materials heavy FTSE 100 has lagged other major indices recently (4% YTD versus +12% for the S&P 500 and +22% for the DAX at the time of writing).

The higher than usual sensitivity to Government policy is a particular worry for investors as political actions are much more dicult to prepare for than company or sector dynamics. The scope for accommodative Government policy that would continue to support Prot Margins is hampered globally by weak Sovereign balance sheets.

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request 8

October 2012

Prot Margins - A Broader Perspective


Portfolio construction and risk management 1. Sectors that have clearly benetted from globalization and where Prot Margins are still higher than their longer term average include Energy, Materials, Capital Goods, Software & Services, Technology and Retailing. Over the medium term, these sectors are at risk from any reversion of secular factors that have driven their Prot Margin growth over the past 30 years. Avoca will be performing further research over the coming months on a sector by sector basis to identify single name investments most exposed to these risks. 2. Over the medium term, investors should focus on sectors that will benet from the growing power of Labour rather than be aected by it. Branded and luxury goods, consumer durables and leisure for example will be long term winners from this trend. Mandelbrot and the long term evidence highlight the potential for Prot Margins to cycle widely around their mean. Prot Margins have not had a serious global dip since the early 1970s and prior to that in the 1930s during the Great Depression. Global energy security and Israel/Iran is an obvious one to watch for as Global Prot Margins are overdue one of these negative fat tail or Black Swan outcomes.

3.

In Avoca, as Value Investors, our focus will always be predominantly on bottom-up, detailed micro-research. Such bottom-up research can however on occasion be very protably augmented by knowledge of historical Prot Margins relative to current levels, particularly on a sector by sector level. Knowledge of long term trends may on occasion help us identify risks that might impact on the Margin of Safety that we require on all investment decisions.

Alan Burke Co-Chief Executive

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

October 2012

Prot Margins - A Broader Perspective


Appendix 1 The derivation of Kaleckis equation
Kalecki assumes intuitively that the economy is made up of households, businesses, governments and the foreign sector or to put it mathematically: Total Output (Y) = Spending by Households (H), Business (B), Government (G) and the Foreign Sector (F) Y= H+B+G+F Equation 1

Note: This is consistent with, but a slightly dierent expression of the well known economic equation that output (Y) is comprised of Consumption (C) (by households and businesses), Investment (I) (by households and businesses), Government spending (G) and Net Exports (NX) more commonly expressed as: Y= C+G+I+NX. If we ignore for the moment any ability to print money or create credit (an important distinction today admittedly), then any decit or excess spending by any of the above categories (H,B,G,F), must be funded by savings from one or all of the other categories. This makes intuitive sense in that for example if the government wants to spend more than it earns, then it will have to borrow that money from the increased saving of one or all of households, businesses or the foreign sector. Another way of stating this is that Savings (NS) for the economy as a whole must equal zero. NS= H(s) + B(s) + G(s) +F(s) = 0 Equation 2

Business Saving B(s) can also be dened as Prots less Investment less Dividends or B(s) = Prots- Investment Dividends Equation 2 above then becomes H(s) + (Prots Investment Dividends) + G(s) +F(s) = 0 Rearranging, this becomes Prots = Investment +Dividends H(s) -G(s) -F(s) Equation 4 Equation 3

The Kalecki equation for the US for 2011 highlights the (intuitive) fact that Corporate Prots are heavily supported today by the level of Government decit spending. Another way of looking at the above is that if the Government decit were to suddenly contract, then US business prots would have to contract unless: households were suddenly to start disaving (i.e. households were to start spending more through consumption or investment); the US current account decit were to disappear (in eect foreigners were to start spending more); or business investment or dividends were to increase (in eect businesses were to start spending more). Again, this is intuitive, and highlights why the equity market may be as focused as it has proven to be on upcoming issues like the US scal cli.

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

10

October 2012

Prot Margins - A Broader Perspective


Disclosure
The information and opinions in this document have been prepared by Avoca Capital Holdings (Avoca Capital). Avoca Capital is involved in businesses and activities that may relate to companies or instruments mentioned in this document. These businesses and activities include, without limitation, trading in various nancial instruments, fund management and investment advisory services. Avoca Capital may trade in the instruments (or related derivatives) that are the subject of this document. Avoca Capital, or funds managed by it, may have a position in the debt of the businesses or instruments discussed in this document. This document is not intended as an oer or solicitation or advertisement for the purchase or sale of any security or nancial instrument. The securities/instruments discussed in this document, if any, may not be suitable for all investors. This document does not constitute investment advice and has been prepared without regard to the individual nancial circumstances and objectives of persons who receive it. Investors should independently evaluate particular investments and strategies and seek their own investment advice. Avoca Capital may not be held responsible for any consequences resulting from any investment in any security/instrument discussed in this document. This document contains information which has been compiled from a number of sources. While the information herein is believed to be reliable, no representation is made by Avoca Capital as to the accuracy or completeness of such information. Nothing contained herein shall constitute any representation or warranty as to future performance. This document is condential and may be neither communicated to any third party nor copied in whole or in part, without the prior written consent of Avoca Capital. Avoca Capital makes no representation concerning the appropriate regulatory treatment, accounting treatment, or possible tax consequences in connection with any investment in the securities/instruments discussed in this document. 2012 Avoca Capital Holdings. All rights reserved

75 St. Stephens Green, Dublin 2, Ireland Phone +353 1 479 3103 Fax + 353 1 479 3129

25th Floor, Heron Tower, 110 Bishopsgate, London EC2N 4AY Phone +44 20 3100 2300 Fax + 44 203 100 2399

www.avocacapital.ie

Avoca Capital Holdings is regulated by the Central Bank of Ireland and subject to limited regulation by the Financial Services Authority. Details about the extent of our authorisation and regulation by the Financial Services Authority are available from us on request

11