#91: SEEDS goes live!


Those of us who see the economy primarily as an energy system rather than a financial one are very much in the minority. Most policymakers and commentators cling to conventional interpretations, even as real events refuse to conform to their world-view. We’re not going to argue our case successfully on theoretical grounds alone, but need evidence to back up our interpretations.

This is what SEEDS – the Surplus Energy Economics Data System – is all about.

The development of SEEDS has been a very big project, almost dauntingly so at times. Now, though, it has reached the point where its output can be made generally available. The aim has been to provide those interested with sufficient data in free-to-download form, whilst not handing comprehensive data free-of-charge to commercial organisations.

Accordingly, SEEDS data has been divided into two products. SEEDS Snapshots are freely available in PDF format, whilst a modest charge will be made for the more comprehensive SEEDS Pro datasets.

I am delighted to inform readers that twenty (out of 22) SEEDS Snapshots are now available for download. You can find them on the resources page newly added for this purpose. This means that you can now access data for Australia, Brazil, Canada, China, France, Germany, Greece, India, Italy, Japan, Mexico, the Netherlands, Norway, Poland, Portugal, Russia, South Africa, Spain, the United Kingdom and the United States. The sets remaining to be added are Saudi Arabia, and the world overview.

After summaries in local currencies and US dollars, the data sheets look first at the energy “mix” for each country – primary energy consumption is broken out into fossil fuels, renewables and an “other” category comprising nuclear and hydroelectricity, whilst production of energy is stated in aggregate.

Next comes a summary of energy economics, including the estimated trend ECoE (energy cost of energy) and EROEI (energy return on energy invested).

Economic output is divided into three categories. The first of these is GDP, stated at constant values. The second, “underlying output”, adjusts GDP for the estimated extent to which borrowed consumption has inflated the headline number. The “real” economy further adjusts the latter for the economic rent exacted by the energy cost of energy. Each of these numbers is then expressed in per capita terms, and rates of growth are stated both in aggregate and at the per capita level.

Further financial data is set out in the remaining tables. Debt at current values is broken out, where possible, into government, household and PNFC (private non-financial corporate) sectors, and the total is also stated in constant, inflation-adjusted terms. Debt is then expressed as a percentage both of GDP and of the borrowing-adjusted underlying equivalent.

Annual growth and borrowing are then compared, in constant terms. Thus, Australian GDP increased by A$39bn in 2015, of which it is estimated that A$30bn was debt-fuelled consumption. Also in 2015, Australia borrowed a net A$284bn, using A$254bn for purposes other than boosting consumption. Over the ten years from 2005 to 2015, each A$1 addition to GDP was accompanied by A$4.35 in net new debt.

The penultimate table summarises government finances in both current and constant values. This is broken out into government revenue, interest paid on government debt, and all other public expenditures, resulting in a surplus or deficit. All of this is set out in current, constant and percentage terms. Thus, Australia’s government deficit in 1995 was A$18bn, equivalent to A$30bn in constant 2015 values, and also equivalent to 3.4% of GDP.

Finally, a similar summary is provided for the external sector. This shows net exports, and the aggregate of current income, which notably includes returns on investments and interest paid on debt. These sum to the current account, a critical indicator of a country’s financial relationship with the rest of the world.

In future articles, we can explore the methods and conclusions of the SEEDS system in depth. For now, though, do download some of these data sheets, and explore what you can get from them.


#90: After peak prosperity


As readers will recall from a previous article, SEEDS – the Surplus Energy Economics Data System – will soon cease to be a purely internal tool, and will be made available to those interested in using it.

As things stand, the intention is to make a summary version (SEEDS Snapshots) available for free download here in PDF format. Professional and business users will be offered a more detailed spreadsheet version (SEEDS Pro) at a modest price. In all, there will be 22 of each – 21 covering individual economies, plus a world economy version.

Rather than issue a technical user manual (though this may yet be necessary), it seems better to introduce the system here using a real example. The balance of reader comments suggests that the subject of this “worked example” should be the United Kingdom. It is hoped that even those readers who are not particularly interested in the UK will find this an interesting example of the economic decline phase after “peak prosperity”.

What follows here, then, is a comprehensive analysis of the British economy, conducted using SEEDS. Conclusions are left to the end, so that readers can follow the process of analysis through from start to finish. Here, for download, is the SEEDS Pro dataset for the United Kingdom. This is the premium version that will be available for purchase after SEEDS goes “live”. It is recommended that you download this now, in order to refer to it during the commentary that follows.

SEEDS 2.15 UK dataset March 2017


Let’s get a couple of technical points clear before we start. First, most economic data is presented in GBP, and the majority of this is expressed at constant values, so that the effects of inflation are excluded. Current data is converted to constant values using the broad-basis GDP deflator index, the base year being 100. (Pending the availability of complete data for last year, 2015 is the base year throughout SEEDS).

Second, where stated in US dollars for comparison, conversion is undertaken using the PPP (purchasing power parity) convention. This is generally superior to conversion using market average exchange rates, though some market-rate data is supplied as well, for those who find it useful.


After some summary tables, the SEEDS analysis begins with two tables related to energy. The first of these (starting at line 44 in the data sheet) is volumetric, and analyses the primary energy position expressed in million tonnes of oil equivalent (mmtoe). Like all tables in SEEDS, this runs from 1980 to 2030, and further amplification is provided by the first chart.

Peak energy production in the UK occurred in 2003, at 272 mmtoe, a number which declined by 59% to just 112 mmtoe in 2015. Against this, consumption has also declined, from 229 mmtoe in 2005 to 188 mmtoe in 2015. This decline reflects a number of factors, including greater energy efficiency, but also the ongoing shrinkage in manufacturing output.

As of 2015, fossil fuels accounted for 83% of British energy consumption, and renewables for 7.6%, which is far higher than the global average for the same year (2.7%). By 2030, the share of UK energy use provided by renewables is projected to reach almost 19%, though this rising share reflects, in part, the downwards trend in aggregate consumption.

The United Kingdom: energy balances


Whether the shift towards renewables has been cost effective is, of course, another question, and plays its part in the table of energy economics (starting at line 56). Obviously, the sharp decline in primary energy production has had a major effect on materials flows and costs. Back in 1999, the UK was a net exporter of 50 mmtoe, or 23% of demand at that time. By 2015, net imports totalled 76 mmtoe, or 40% of consumption.

This changing material balance has necessarily impacted the UK’s energy costs, measured in the data sheet both as ECoE (the energy cost of energy) and as EROEI (the energy return on energy invested). Reflecting global trends, the estimated ECoE of consumption (line 61) has risen from 3.8% in 2000 to 7.8% in 2015, which is not significantly different from a world average of 7.0%.

But the swing from net exporter to net importer has had a dramatic effect on the overall ECoE of the economy (line 62). Being a net exporter is advantageous, mainly because costs and taxes (which, of course, are revenues for suppliers and the government) are incurred at home rather than overseas. Accordingly, Britain’s overall ECoE is estimated to have soared from just 0.9% in 2005 to 10.3% in 2015. As we shall see, this has had a major adverse effect on prosperity.

For those who prefer EROEI measures, that of the UK in 2015 is put at 10.7:1, and is projected to fall to just 6.6:1 by 2025. Readers who understand EROEI will appreciate that a ratio this low poses a dire threat, not just to prosperity but to economic viability itself.

The United Kingdom: energy economics


The role of borrowing

The next table to look at in the data sheet considers growth and borrowing, and starts at line 125. Expressed at constant 2015 values, annual growth in GDP ranged between +£48bn and -£72bn between 2005 and 2015. The total of growth over this period was £215bn, which lifted GDP (line 69) from £1,649bn in 2005 to £1,864bn in 2015.

Over the same period, however, total debt (line 114) increased from £3,580bn to £4,950bn, a rise of £1,369bn. The trailing ten-year (T10Y) relationship between borrowing and growth was, therefore, 6.37:1, meaning that £6.37 of borrowing accompanied each £1 of recorded growth in GDP.

The relationship between annual growth and borrowing can be seen in the next chart.

The United Kingdom: growth and borrowing


Clearly, borrowing on this scale is likely to have inflated GDP, by funding present consumption at the expense of future liabilities. What this really means is that, without this borrowing, growth would have been lower.

But how much lower? According to the formula used by SEEDS to measure this effect, £182bn of borrowed money was used to finance consumption between 2005 and 2015.

Though this estimate equates to just 13% of total borrowing (of £1,369bn) over that period, and is probably conservative, it accounts for most (85%) of growth recorded between 2005 and 2015. Reference to line 70 of the datasheet shows that, adjusted for the effect of borrowed consumption, underlying GDP in 2015 is estimated at £1,467bn.

This in turn means that, without borrowed consumption, GDP in 2015 would have been 21% lower than the reported number. It also indicates that trend growth is just 0.4% (line 79) rather than the generally-assumed c2%. On a per-capita basis, underlying growth is negative 0.3% (line 94), because the population is growing more rapidly than underlying (borrowing-adjusted) GDP. Even this, of course, is before we take trend ECoE into account.

The next chart shows these trends at a glance. The blue line is reported GDP, including consensus expectations out to 2021. The black line is GDP adjusted to exclude the impact of debt-funded consumption. The red line is the real economy on an ex-ECoE basis, and is closely analogous to prosperity as individuals experience it.

The United Kingdom: economic output


Debt dangers

If we now turn to debt aggregates (starting at line 105), it will be seen that constant, inflation-adjusted debt increased from £3,580bn in 2005 to £4,950bn in 2015 (line 114). Debt trajectories after 2015 are projected using an algorithm, and this estimates end-2016 debt at £5,196bn.

This is almost certainly a serious under-statement, as data for the first nine months of 2016 show that debt actually climbed to £5,408bn in that period, a far larger increase (of £458bn) than the SEEDS algorithm estimates for the whole of the year (£247bn).

If something like this is confirmed by final data, future debt projections will need to be revised upwards. Even as things stand, the ratio of debt to GDP, having risen from 209% in 2005 to 265% in 2015 (line 122), is projected to reach 300% by the end of 2019.

Of course, this ratio refers to reported GDP – were the underlying (ex-borrowing number) used instead, the ratio already exceeds 330%, and will be well over 400% by the end of the decade (line 123).

The United Kingdom: debt


The table of external flows (starting at line 157) reflects a steady deterioration in the current account, from a deficit of 1.2% of GDP in 2005 to 4.3% in 2015. This is not a reflection of net trade, which has in fact been on an improving trend. Rather, income (which primarily comprises net returns on equity and debt capital) has swung from +2.1% of GDP in 2005 to -3.3% in 2015. At constant values, this swing equates to £96bn (from +£35bn to -£61bn) (line 169), or 5.1% of current GDP, over a ten-year period.

Finally, we need to factor ECoE into the equation in order to measure “real” or “discretionary” income. This measure which has a direct impact on perceived prosperity, because it impacts the income that households have left after the cost of essentials.

On an aggregate basis, the real economy was 7% smaller in 2015 than in 2005, and SEEDS indicates that Britain hit “peak prosperity” back in 2003 (line 71). On a per-capita basis, discretionary income declined by 13% between 2005 and 2015 (line 87), and is falling at a trend rate of about 1.4% annually.


Overall, then, the SEEDS assessment of the British economy is very bearish. Individual prosperity is deteriorating – as is the aggregate, once the effect of “borrowed consumption” is adjusted out – whilst debt continues to rise markedly. Dependency on overseas creditors has become acute, mainly because income flows have been impaired by past patterns of asset sales and borrowing from abroad.

Looking ahead, the deterioration in British economic performance is starting to look irreversible, and certainly cannot be halted, let alone reversed, without wholesale changes in policy.

* * *

Here is the new PDF for Canada:

2.02317 Canada

#89: Chinese whispers


Which is the world’s largest economy? Converted at market exchange rates, China ($11tn) is smaller than the United States ($18tn) but, on the PPP (purchasing power parity) basis of conversion widely regarded as superior, China (at $21tn) now takes the top spot.

Be that as it may, there can be little doubt that the economy of China is second to none in terms of its importance to global growth. In local currency, Chinese GDP was RMB 68tn in 2015, four times larger in inflation-adjusted terms than it was in 2000. China has been the primary engine of world growth since the millennium, and most observers, it seems, expect it to remain so for the foreseeable future.

In short, if the Chinese economy were to catch a cold, the world economy would be in for a bout of influenza at best, and could well face the economic equivalent of pneumonia. This is as true of debt as it is of growth – wobbles in China would trigger shock-waves around the world.

There could not, then, be a more important subject than China for evaluation using the Surplus Energy Economics Data System (SEEDS).

Readers will find a downloadable PDF of SEEDS statistics for China at the end of this article.


For China’s army of Western admirers, the conclusions set out here are disturbing.

Most strikingly, Chinese economic growth has, over the last decade, become a hostage to borrowing on a gigantic scale. Though used primarily for capacity expansion rather than for fuelling consumption, this borrowing has had a hugely distorting effect on growth. Were China to cease borrowing about 20% of GDP annually, as she does at the moment, growth would fall back from 6.5% to a trend rate of 3.4%.

There must be limits to quite how long China can go on using borrowing to create growth. Based on reported GDP, debt already stands at 246% of reported GDP, up from 141% just seven years ago. When measured against an underlying GDP figure stripped of estimated debt-funded consumption, the ratio climbs to 384%. On this underlying basis of measurement, the ratio could hit 500% within just five years.

In short, the Chinese economy is following the tried-and-failed Western policy of using debt to manufacture “growth”.

There are clear limits to how much longer she can go on doing this.

GDP and debt

It must be understood from the outset that the reliability of much Chinese data seems questionable. At the simplest level, key metrics like growth and unemployment appear not only remarkably unvarying but are, equally remarkably, always exactly in line with official expectations. Some economic numbers seem hard to reconcile with non-financial, volume indicators.

There are also issues of discontinuity, where methods of calculation seem to change without the data for prior years being restated to match. It should also be understood that international data sources largely replicate these issues, since they are necessarily based on Chinese official statistics.

Let’s start with an indicator always regarded as pivotal in SEEDS analysis – the relationship between GDP and growth on the one hand and, on the other, debt and borrowing. The first chart compares GDP and debt in local currency at constant (2015) values. (Debt numbers used throughout this analysis exclude the inter-bank or “financial” sector which, were it included, would probably lift total debt from RMB 168tn to somewhere nearer RMB 200tn).

1. China: GDP and debt, 1995-2021F

China bespoke 1 GDP & debtjpg_Page1

As the chart shows, a remarkable divergence has emerged between debt and GDP in the years since the global financial crisis (GFC). Between 2007 and 2015, and expressed at constant 2015 values, debt increased by 228%, from RMB 51tn to RMB 169tn. This far outpaced expansion in output, where reported GDP grew by 154%, to RMB 68tn in 2015 from RMB 35tn in 2007.

Borrowing and growth

Of course, percentage changes are important, but what really matters is the relationship between borrowing and growth. This is summarised in the next chart, which shows annual borrowing and growth as percentages of GDP.

2. China – borrowing and growth, 2000-2021F

China bespoke 2 B & Gjpg_Page1

The picture that emerges is quite extraordinary. Over the ten years between 2005 and 2015, GDP grew at rates of between 9% and 14% annually, not even stumbling materially during the 2009 global downturn. But debt has grown by between 17% and 35% of GDP each year, with the exception of 2009, when debt increased by 47% of GDP.

What this means is that, over a period in which reported GDP increased by RMB 40tn, debt expanded by RMB 129tn. This is a borrowing-to-growth ratio of 3.2:1, still reasonably modest by Western standards, but a far cry from past Chinese practice – back in 2005, the trailing ten-year (T10Y) ratio was only 1.67:1.

The Chinese debt model

This pace of change in the trailing average means that the relationship between borrowing and growth merits close consideration. China retains comparatively low levels of household indebtedness, which stood at 39% of GDP in 2015, compared with 79% in the United States, 87% in the United Kingdom and 104% in the Euro area. Households accounted for a modest 18% of all net borrowing in China between 2005 and 2015. Government debt, too, remains low by world standards, at 44% of GDP. Neither households nor the state, then, are bingeing on borrowed money.

But the biggest share of Chinese borrowing by a wide margin is the PNFC (private non-financial corporate) sector, which borrowed RMB 83tn in the decade to 2015, or 64% of all borrowing. This increased total PNFC debt to RMB 112bn in 2015, from about RMB 29tn in 2005, an increase so huge that readers should be reminded that these are constant numbers, adjusted to exclude inflation.

Unlike the Western economies, whose vice-of-choice is to use debt to fund consumption and inflate property markets, the Chinese bias is towards using debt for investment in capacity. In theory, capacity investment should be “self-liquidating”, because capacity increases should increase income, and thus fund the paying off of the initial debt. (This is contradistinction to consumer borrowing, which is “non-self-liquidating”).

But the self-liquidating characteristic of business investment depends on capacity expanding without depressing margins, something which happens when expansion creates major capacity surpluses. It is abundantly clear that Chinese PNFC borrowing has followed the course of excess, depressing returns in the process.

As a result, much of the Chinese business sector earns returns which appear to be well below the cost of debt capital. In this situation, an obvious remedy is to convert debt into equity. This, however, seems to have been tried, and failed, because it showed clear tendencies to crash the equity market.

Of course, PNFC borrowing differs from consumer borrowing, but it nevertheless finds its way into consumption and economic activity. If a business invests in new capacity, much of that investment finds its way into the pockets of households, either directly, through employment, or via suppliers. High investment helps diminish unemployment but, where this investment is non-self-liquidating, the effect is “borrowing to employ”, and this has parallels with Western-style “borrowing to consume”.

The judgment call here is this – how much Chinese borrowing has inflated consumption (and broader activity) artificially, meaning that this activity would disappear were borrowing to cease? The calculation made by SEEDS ascribes only 16% of all borrowing between 2005 and 2015 to consumption, probably a conservative estimate of how much net new debt has been used to inflate activity. Even so, this amounts to RMB 20.8tn, equivalent to slightly more than half of all growth (of RMB 41tn) in GDP over that period.

If that estimate is correct, underlying growth in Chinese GDP over the last decade and more has been far lower than the reported numbers.

As the next chart shows, this calculation puts Chinese aggregate growth since 2000 at 160% (RMB 27tnb) – still impressive, but nowhere near the reported 295% (RMB 51tn). It indicates that underlying GDP in 2015 was around RMB 44tn, a very long way adrift of the reported RMB 68tn.

3. China – reported and underlying GDP, 1980 – 2030F

China bespoke 3 GDP % ULjpg_Page1

For how much longer?

The implications of this assessment are stark.

First, SEEDS estimates current trend growth at around 3.4%, far below official rates of 6.5%. It also trims per capita trend growth to 2.9% from 6.0%.

Of course, China can still record growth rates at or above 6% – but only if the country continues to borrow at recent levels. That would mean adding yet more surplus capacity, and depressing margins still further.

The final sting in the tail of this analysis is that, if underlying GDP is a lot lower when stripped of the borrowing effect, debt ratios are correspondingly higher. On the SEEDS basis of computation, aggregate debt already stands at 385% of GDP (rather than the reported 246%), and is growing a lot more quickly than publicly available numbers indicate, adding around 43% of GDP (rather than 20%) annually.

With the export-based model faltering, and with a great deal of economic activity dependent on borrowing, China may have ceased to be the powerful engine of growth that is so customarily assumed.

4.China – debt/GDP ratios, 1995-2021F

China bespoke 4 debt ratiosjpg_Page1

China PDF

#88: SEEDS to go live?


In recent times, it’s fair to say that three broad themes have dominated discussion here.

The first of these, of course, is surplus energy economics – the philosophy which says that the economy is an energy system, not a financial one. Money may be the map of the territory, but the territory itself is energy.

The second “hot” topic is Ponzi finance. In the early years after the millennium – and for reasons which the surplus energy interpretation can alone explain – real economic growth petered out. Since then we’ve been faking it, spending borrowed money and calling this “growth”. All the while, of course, debt (and informal “quasi-debt”) has escalated. Essentially, the powers that be have been busy destroying the future, not just by accumulating debt but also by crippling other forms of provision for the future, most obviously pension funds and other forms of saving.

Third, the general public has started to smell a rat. I don’t mean that the public understand surplus energy theory, or spend their time comparing growth with borrowing. But the public does have an intuitive sense of when things are going wrong, and this is one of the reasons why they are busy repudiating the “liberal” elites, along with much of what these elites stand for.

The bottom line of these three themes is that policymakers and economists – as some of the latter, to their credit, acknowledge – don’t understand what is happening. To tackle this, we are in urgent need of new economic understanding.

This is urgent, because what the authorities have been doing for a decade and more has been akin to carrying out brain surgery with carpenters’ tools. They can’t fix the economy because they don’t really understand how it works.

And this is where SEEDS comes in. To explain why, I need to digress a bit.


After publishing Perfect Storm (when I was head of research at Tullett Prebon), I embarked on writing what was to become Life After Growth. Throughout that project, I realised that there was a glaring need for a comprehensive mathematical insight into the energy economy. For the book, I was able to include general trends in ECoE (the energy cost of energy), and had data to illustrate, for example, the progression of energy, population and economic output over time. What I did not have, however, was a level of granularity enabling evaluation of individual economies – and the world economy as a whole – with any level of detail.

This was why SEEDS – the Surplus Energy Economics Data System – was developed. Initially, the aim was to estimate ECoEs across fuel sources and economies. At first, SEEDS simply provided estimates of ECoEs on a country-by-country basis over time. The thinking was that, once we had a grasp of ECoEs, we could deduct them from GDP to work out what the “real” economy of goods and services was really doing behind the public facade of recorded output. This has been developed to the point where twenty-one countries are covered by the system – these are the United States, Canada, Mexico, Brazil, France, Germany, Greece, Italy, the Netherlands, Norway, Poland, Portugal, Russia, Spain, the United Kingdom, Australia, China, India, Japan, Saudi Arabia and South Africa.

At the same time, it became apparent that a lot more economic content was required. The “wish list” at this point included debt and borrowing, trade and external flows, government finances, and measures on a per capita basis. Above all, the aim was to get into the dynamics of borrowing and growth, and pull all of these together with the fundamentals of ECoE and surplus energy.

With all of this accomplished and just a few tweaks remaining after several years of effort, the latest version – SEEDS 17 – has become an extremely valuable interpretive tool. To use a motoring analogy, SEEDS has evolved from a Morris Minor into something reasonably akin to Mercedes.

But even the best car achieves nothing whilst it remains in the garage.

Going public?

The next objective, logically, is to make SEEDS generally available.

Broadly, there are two main “audiences” for SEEDS. The first are individuals concerned about what is happening, and keen to further their understanding. The second are professionals, engaged either in making decisions or in providing advice. (There is actually a third category, comprising academics and non-profit organisations, but how to meet their requirements remains to be decided).

This argues for two kinds of product. The current plan is that, for the general public, comprehensive data will be available as PDFs which can be downloaded free of charge. The working name for these is SEEDS Snapshots. For professionals, a more detailed data package (known as SEEDS Pro) will be made available at what I hope will be a pretty reasonable price.

This project is still under development, and it may be some while before SEEDS Snapshots and SEEDS Pro go live. Also, as you may know, I’m planning a sequel to Life After Growth.

Meanwhile, of course, I’ll continue to post articles here – and please do keep making your very helpful comments.