Do Energy Saving Light Bulbs Really Save Energy?

Some considerations about the usefulness of energy saving light bulbs.

Conventional light bulbs which are now banned in the EU since 1st September 2012 are very inefficient when it comes to turning electrical energy into visible light. More than 90 % of the energy input are emitted in the form of heat. Thus with a less than 10 % efficiency in terms of light production the classical light bulb may indeed look a rather hopeless case and energy saving lamps appear to be the preferable choice.

However, a closer inspection shows that the odds are not at all so bad for the conventional lighting medium. During the cold season, i.e. whenever people feel the need of switching on their heating, the classical light bulb contributes to the heating effort. Thus whenever we count heading degree days the old fashioned light bulb lowers the need to switch on the heating. From a physical point of view during the heating period the energy saving light bulb does not save energy. That is a simple deduction from the law of energy conservation. On the contrary, if we turn to energy saving lamps, the work load of our heating systems gets higher.

On the other hand, it is true that during the warmer period energy saving light bulbs are the better choice because then we really do not need any extra heating. So in order to understand the energy balance of both lighting systems we developed a model which is based on the following assumptions:

Get-up time: 6:00

Leaving home in the morning: 8:00

Returning home in the evening: 17:00

Sleepy time:  24:00

In addition to that we assume an average performance of 200 W(el). Our model is then applied to two European cities, one in the northern (Stockholm) and the other in the southern part (Rome). From this model it follows that the maximum number of lighting hours per day is 9 which is reached during the dark winter months. On the other hand, during the long daylight hours in summer, it may be necessary to switch on the light for no more than 2 hours per day. Electricity consumption for light sources is thus fluctuating between those two extremes. Fig. 1 shows the distribution of lighting hours in both cities.

Fig. 1 Lighting hours in Rome and Stockholm.

How much artificial light we actually need is largely determined by the length of the daylight period which in turn is governed by sunrise and sunset.  The data for those astronomical observables are easily accessible, clearly highlighting the differences between northern and southern locations. Although these differences may be very large during certain periods of the year, their overall impact on our model is far less dramatic than expected. On an annual basis the number of lighting hours in Stockholm is not very much different from the one in Rome, 2160 vs. 2110, respectively, i.e. less than 3 percent difference. These lighting hours correspond to an average electricity consumption of 432 kWh and 423 kWh, respectively.

Lighting and heating

What is, however, different is the number of heating degree days between the two cities. Correspondingly, the number of lighting hours during the heating period is much larger in northern Europe than in the southern part. Our analysis shows that more than 80 % of lighting hours are consumed during the heating period in the case of Stockholm. This, in turn, means that during that period there is no gain from using energy saving light bulbs. The respective number for Rome is 62 %. Thus also in southern Europe a substantial amount of lighting is used when people are likely to put on their heating. The rest of Europe lies, in its vast majority, somewhere between these two extreme values. There are exceptions, but these are statistically insignificant.

From a physical point of view, energy saving lamps are only useful when it is warm outside, because then there is no need for excessive heat. During winter those lamps do not lead to an overall reduction of energy consumption. Let us look at the situation from the point of view of a classical light bulb. It wastes energy in summer, but not in winter. According to our model, the waste energy produced by a classical lighting system amounts to about 78 kWh (Stockholm) and 145 kWh (Rome), respectively, over the whole year. We may also ask how much heating energy we can save by using normal light bulbs. The figures are 311 kWh (S) and 235 kWh (R), respectively. Thus we may conclude that, physically speaking, the energy balance of classical light bulbs is clearly better than the one of their energy saving competitors. Their contribution to heating in winter outweighs their waste of energy in summer. Of course, the effect is much more pronounced in northern Europe than in the southern countries. But even in the latter ones, the result is undoubtedly in favour of normal light bulbs. Figs. 2 and 3 demonstrate the contribution to heating and the waste energy of conventional light bulbs for the two cities.

Fig. 2 Heating potential vs waste energy of conventional light bulbs in Stockholm.

Fig. 3 Heating potential vs. waste energy of conventional light bulbs in Rome.

It is, of course, conceivable that under certain circumstances energy saving light bulbs live up to their expectations. Our model indicates that this may be the case in tropical areas where heating is hardly ever needed.

In any case, energy-saving lighting systems are not reducing energy consumption at the level of private consumption which we have considered inour model. This is even less so at a more general level, when taking into account the energy effort for producing and disposing of light bulbs. The recycling of energy saving systems is putting an extra burden on their overall energy balance, therby seriously questioning their high expectations.

Energy Efficiency – How Europe Can Achieve Its 2020 Targets

Becoming more energy efficient is perhaps the most straightforward and least expensive way of tackling the energy problem. Recently, the EU has addressed this issue within the framework of the so-called Europe 2020 targets which aim at reducing gross energy consumption by 20 %, producing at least 20 % of all energy from renewable sources and reducing greenhouse gas emissions by 20 % (with respect to 1990 levels). All that is supposed to be attained by 2020 at the latest.

Leaving aside the two latter issues, we will focus in this posting on the question of energy efficiency. Lowering energy consumption by 20% (when compared to projected levels) means in particular that by 2020 Europe will use some 1474 Mtoe (mega-tonnes of oil equivalent) of primary energy.

Fig. 1 shows the development of EU gross inland consumption from 1990 onwards with EU-27 representing the entire union whereas EU-15 refers to the “old” Member States, i.e. excluding those countries which joined the union in 2004 or later.  One striking observation is that during the past two decades consumption figures have always been substantially higher than the 2020 target line. Thus we are facing a real challenge.

Fig.1 EU gross inland consumption of energy. Source: Eurostat.

But looking at absolute consumption levels only does not reveal the whole story since, at the same time, we are also expecting economic growth. And a growing economy means higher energy consumption, at least to some extent. Putting consumption and economic performance together yields another interesting observable, namely the so-called energy intensity which is shown in Fig. 2. This parameter indicates how much energy is needed in order to produce one unit of economic output. Energy intensity is thus measured in kgoe/kEUR (kg of oil equivalent per 1000 EUR). Apparently, this indicator has fallen drastically since 1991.  In 2010 it was at 168 kgoe/kEUR for EU-27.

Fig. 2 EU energy intensity. Source: Eurostat.

One apparent feature of this figure is that the gap in the intensity levels between EU-27 and EU-15 is getting smaller over the years, thus indicating that the countries which joined the EU in 2004 or later are outperforming the older Member States (EU-15) when it comes to becoming more energy efficient. Nevertheless, the energy intensity of the younger EU members is still considerably above average.

Reducing absolute energy consumption means that intensity figures will drop accordingly. But by how much? In order to obtain an answer to this question, we analysed two scenarios, one with a stagnant economy, i.e. no (real) GDP growth up till 2020, and another one with an average GDP increase of 2 % annually.

Taking the zero-increase economy as a reference we find that energy intensity must drop from its 2010 level to some 141 kgoe/kEUR in 2020. This is not too far from the current EU-15 level (151 kgoe/kEUR). However, at EU-27 level this means that the intensity has to go down by some -1.75 % on average per year.

Going over to a more dynamic scenario with an average economic growth rate of 2 % we find the respective energy intensity in 2020 at 115.5 kgoe/kEUR. Obviously, the effort is much stronger in this case, requiring an annual decrease of almost -3.7 %.

To put things into perspective we may mention that the average intensity gain during the period 1991-2010 was 1.94 % per year. Thus, the prospect of performing equally well in a no-growth economy does indeed look quite promising. However, once the economy is supposed to grow even at a moderate pace, our effort may easily double.  In that case, more drastic measures are required in order to attain the ambitious goal.

Does Saving Energy Push Renewables?

Yes it does. Let us look at a concrete example in order to get the point. The EU plans to improve its energy efficiency by 20% by 2020. In other words, 20% less energy will be used by then, according to plans. The baseline is the primary energy consumption for 2010 which was 1770 Mtoe. Thus, if all measures are in place, by 2020 this figure should be down to 1416 Mtoe.

In all likelihood, the savings will concern almost exclusively the use of conventional energies (coal, nuclear, oil) whereas renewable energies will not be touched by this development. Therefore, we may safely assume that on the consumption side renewables will be equally well off  as they are now. In fact, this is a very conservative estimate. On the contrary, renewable energy use may well be expected to rise over the next decade. But let us stick to our conservative approach for the time being. In 2010, the consumption of renewables amounted to some 172 Mtoe corresponding to 9.7% of total consumption.

Fig. 1 EU Gross inland consumption in 2010

Given our  2020 scenario from above and keeping renewable consumption at 172 Mtoe, we may conclude that by then renewables account for about 12.2% of total consumption. Bear in mind that this is true even if energy production from renewable sources does not increase.

The projection for 2020 would consequently look like this.

Fig. 2 EU Gross inland consumption in 2020

Thus, by saving energy the relative weight of renewables in the energy mix is automatically increased. The bigger the savings on the one hand the bigger the extra share of renewable energies on the other.



Heating Degree Days and Energy Consumption

Heating degree days (HDD) may serve as an indicator for the amount of energy used for heating purposes. The correlation seems to be pretty obvious: a larger number of HDD should inevitably lead to a corresponding increase in energy consumption. This relationship should, as a consequence, be reflected by the amount of primary energy used. Of course, heating is not the only way to consume energy. Traffic, industrial production and services equally request their share in primary energy demand.

In Germany, heating accounts for about 30 % of total final energy consumption. Thus, if the number of HDD is up by, say, 10 % then we would expect the consumption figures to increase accordingly. The question is to what extent the latter would reflect changes in HDD. Let us demonstrate this via a simple thought experiment. Imagine Germany consumed 100 units of final energy in 2009, 30 of which were used for heating. The number of HDD was x. In 2010 HDD increased by 10% compared to the previous year. Thus we would expect a total of 33 units being absorbed for thermal comfort. Everything else remaining unchanged, the total final energy consumption in 2010 would amount to 103 units. Thus, the total consumption figure would be up by 3% in 2010.

Does this argument also hold good for primary energy consumption? Let us have a look at two countries of similar size and climatic conditions, namely Germany (Fig. 1) and UK (Fig. 2). The source data have been taken from BP Statistical Review of World Energy 2011 and Eurostat. The figures show primary energy consumption per capita (CPC) in tons of oil equivalent (toe) and the number of HDD in the respective country.

Fig. 1 Primary energy consumption per capita (in toe) and HDD in Germany

Consumption figures in Germany reflect changes in HDD only partially, as expected. At the end of our observation period we even note that a significant rise in HDD is met by a slump in consumption per capita. Between 1994 and 1996 HDD went up by more than 27%. The respective rise in CPC was a mere 3.3%.

The UK data are as follows.

Fig. 2 Primary energy consumption per capita (in toe) and HDD in UK

Again, the changes in consumption per capita are much less pronounced than the respective variations in HDD. In 1995/96 HDD increased by some 11.5 %, whereas CPC went up by 4.5 % only. As in the German case, towards the end of the obervation period a clear upward trend in HDD is met by a significant drop in CPC.

Heating degree days have certainly their merits when it comes to estimating energy needs for thermal comfort. However, on a more global scale, their usefulness is relatively limited. In any case, their importance should not be overrated.

Heating degree-days

What are heating degree-days and what are the advantages and limitations of that conept? Generally speaking, heating degree-days (HDD) represent a sensible measure in order to estimate how much energy must be provided for heating purposes.

It´s cold outside, you turn on the heating. The colder it is, the more you have to heat, if you want to keep the room temperature at a convenient level. As a consequence, you need more energy, if the outside temperatures are lower. Thus the temperature difference between inside and outside to a large extent determines how much oil, gas, wood or electricty you need in order to keep your place cosy and warm.

But this is not the only parameter having an impact on your heating bill. Another factor of crucial importance is the numer of days you have to keep the heating running in the first place. Wintry weather conditions and their duration can vary considerably from one year to another. Last year, at the beginning of November, outside temperatures in northern Europe were already below 0° C. This year, however, in the same region the thermometer has hardly ever touched the freezing point, thus saving a lot of energy costs.

So we see that two crucial parameters determine the heating effort: the temperature difference between living room and outside on the one hand and the duration of the period when the heating is on.

Formally speaking, following the definition used by Eurostat, HDD may be defined as follows:

HDD = (18° C – Tm)*d,  if Tm <= 15° C  or

HDD = 0,  if Tm > 15° C

In this formula, d represents the number of days when heating is considered to be required and Tm is the mean outside temperature defined as Tm = (Tmin + Tmax)/2. Thus, Tm is an average value of minimum and maximum temperatures during a certain period. But when is the heating actually on? According to Eurostat the heating is on when Tm <= 15° C, whereas for Tm > 15° C it is off and then HDD = 0.

In this way, we have obtained an important indicator for the amount of energy which is needed in order to keep our living or working space at an agreable level.

However, HDD in itself is not sufficient to determine or even estimate the actual amount of energy necessary for heating purposes. To that end, more input is needed. In particular, we need to know how big the energy flow from our living and/or working premises to the outside world is. Clearly the heat flow is directly proportional to the difference in temperatures as indicated in the formula for HDD. Yet, the amount of heat passing from the cosy appartment to the cold and sometimes frosty environment also depends on the insulation we use in order to reduce the loss of heat. The insulation in turn is closely linked to the construction materials used.

Fig. 1 gives us an overview over HDD in the EU-27 and some selected countries. The raw data for this have been taken from Eurostat.

Fig. 1 HDD in EU-27 and selected countries, 1980-2009

Apparently, there is a clear distinction between several countries, depending on their geographical location. HDD for Germany and UK are closely following the EU average. The northern countries Sweden and Finland are placed well above that average, whereas the southern Member States Spain and Portugal find themselves well below that value. The mean deviation from the EU average amounts in the case of Sweden and Finland to 67% and 79%, respectively. Spain and Portugal, as the antipodes in HDD,  are as far as 43% and 60% below the European mean value, respectively.

HDD reflects the climatic conditions of each country. Average temperatures are considerably lower in Europe´s northern periphery and in the southern part. Therefore the difference in HDD between Finland (5800 on average) and Portugal (1300) is easily explained. Taking HDD as the only reference, Finland would need more than 4 times as much energy for heating than its couterpart. However, comparing these figures with the energy consumption per capita for both countries (Finland 5.3 ktoe and Portugal 2.2 ktoe, annual average for 1991-2010) yields a clear indication that there must be some features which tend to soften the sharp discrepancies. Among these are the standards for heat insulation (which can vary between different countries), the number of cooling degree days (having an opposite north-south tendency) and the level of industrialization.

Oil Dependency of Developed Economies

Oil is one of the major energy sources for a modern economy. Both, developed and developing economies depend heavily on it. So we may ask ourselves to what exent we depend on this critical source. Intuitively, we know that renewables are constantly gaining ground. However, the simple fact that oil prices continue to be a vital indicator for economic activity shows us that oil still keeps its dominant role in the energy mix.

In order to find out how our dependency on oil and oil products has developed over the past decade, we compare the economic output in terms of nominal GDP with the respective oil consumption figures. This is done for the EU, the United States and Japan. The period in question is running from 2000 to 2010. Both, the GDP and oil consumption are normalized to be equal to 100 in 2000. The raw data for our investigation have been taken from Eurostat and the Shell Statistical Review of World Energy 2011.

Let us start with the European Union. Fig. 1 gives us a nice impression about the decoupling of economic activity and oil consumption which has taken place in the past decade. A net gain in real GDP is accompanied by a significant drop in oil use.

Fig. 1 EU-27 oil dependency 2000-2010, 2000 = 100.

The underlying reasons for this significant development are twofold: on the one hand, oil is facing competition from other sources such as natural gas. On the other hand, oil using machinery, like car engines etc. are getting more efficient, i.e. using less energy per km/mile.

Fig. 2 displays the same analysis for the United States. Again, real GDP and consumption of oil are jeading in different directions. As in the case of EU-27, the decoupling becomes even more siginificant as of 2006/2007. Quite remarkably, during the economic crisis in 2008/2009 the relative drop in consumption was considerably bigger than the one in economic performance.

Fig. 2 US oil dependency 2000-2010, 2000 = 100

As a final example, let us have a look at the situation in Japan. In one of our previous post we have already observed that Japan excels particularly when it comes to energy intensity, i.e. economic output per unit of energy used. Having this in mind, we would expect quite similar findings for the case of oil consumption. Fig. 3 shows the results of our analysis.

Fig. 3 Japan´s oil dependency 2000-2010, 2000 = 100

Although Japan´s GDP has performed less favourably when compared to the US and the European Union, its oil dependency has fallen much stronger than the one of its competitors. The decoupling between economic performance and the respective oil consumption is already quite significant in the beginning of our observation period, getting larger during the years. Thus, the reduced consumption of oil and its products is one of the key factors in Japan´s successful struggle to obtain a higher economic output per unit of energy.

Energy Intensity in Europe, the US and Japan

In the previous posting we analyzed the development of energy intensity at a European scale. The findings were twofold: on the one hand, we saw a clear tendency to lowering the amount of energy per unit of GDP. This means that energy is used in a more efficient way. On the other hand, there are still remarkable differences between the EU member states. The gap between, say, Spain and Denmark which amounted to 63.72 kgoe/1000 EUR in 1995 has actually widened over the years and was at 79.44 in 2009. Thus, Denmark has clearly done better than Spain during that period. This, in turn means, that there is substantial room for improvement on the Spanish side.

Arguably one might say, that Spain and Denmark are not at the same level in terms of productivity, and that is certainly a valid point. However, from the Spanish point of view it is strongly desirable to become more competitive and thus increase its productivity.

In this post we want to have a closer look at the energy intensity of the three main economies in the world having comparable levels of productivity: the EU, the US and Japan. The raw data for the following analysis have been taken from Eurostat. As usual, the quantity in question is measured in kgoe/1000 EUR of GDP.

Fig. 1 Energy intensity in the EU, US and Japan, kgoe/1000 EUR

First, we observe a decline of energy intensity in all three economies. However, this decline is much more pronounced in the EU and the US than in Japan. During the period in question the US saw its intensity figure falling by 25.6%, while Europe faced a decline of 20.9%. Japan, on the other hand, came down by a mere 11.8%. Why is that so? It seems that Japan has already reached a saturation level when it comes to using energy in the most efficient way. The US and Europe have considerably improved their output figures, delivering a higher GDP per unit of energy used.

Nevertheless, there is still a huge gap between the two “Western” economies and Japan. Clearly, the gap is narrowing. In 1995, it was some 104.9 kgoe/1000 EUR between the EU and Japan, while the respective difference between the US and Japan was 134.6. In 2009, this has come down to 73.5 (EU-Japan) and 85.7 (US-Japan), respectively. Thus, the United States are still using almost twice as much energy per unit of GDP as Japan.

Improving productivity and introducing energy saving measures are the key parameters if we want to perform equally well as Japan. Clearly, Japanese economy has set the baseline which we should try to achieve. It is possible to bring energy intensity down to less than 100 kgoe/1000 EUR. However, this may take several decades given the current level of progress.

Energy Intensity

Common opinion holds that if economic activity is increasing the consumption of energy will follow suit. At first glance this seems a convincing argument: producing 10 cars uses 5 times more energy than producing 2 cars. However, reality is not quite that simple.

First of all, there are scale effects coming into play. You do not switch on the whole production chain for each car individually, but rather try to produce the whole lot “in one go” which means that the entire production process will become more efficient which, in turn, helps saving energy. This essentially means that the scaling factor in the above example is no longer 5 but less than that.

Apart from making economic processes more efficient there are other factors which determine the level of energy intensity. Introducing energy saving measures, using machinery with a lower energy consumption, changing consumption patterns and other issues may lead to a lower energy intensity. So what is energy intensity? It is defined as the inland consumption of energy (coal, oil, gas, electricity and renewables) per unit of GDP within a certain period, usually one year.

As energy is an important cost factor it is desireable to minimize its use per economic output. This is true not only at the level of entreprises or businesses, but also for the economy as a whole. If we manage to produce more with the same (or even less) energy we may in the long run reduce our import dependency. However, so far the successful lowering of the energy intensity at European level has not yet led to a significant reduction of energy imports from third countries.

In Fig. 1 we see the energy intensity in kg of oil equivalent (kgoe) per 1000 EUR of GDP for EU-27, EU-15, Germany, France, Italy, UK and Spain from 1990 to 2009. Note that there are no data available for 1990 for EU-27. EU-15 and Germany (year of unification between East and West Germany). All data are taken from Eurostat.

Fig. 1 Energy intensity in kgoe/1000 EUR of GDP

One striking observation is that all countries of our selection as well as the EU as a whole have managed to reduce their energy intensity considerably since 1995. At EU-27 level the respective level has gone down by almost 21 %. Germany could reduce its energy intensity by almost 18 %, whereas UK managed to cut it by more than 30 %. On the other hand, the figures for the southern countries Italy and Spain are less impressive with 6.8% and 6.1%, respectively.

Another remarkable feature of our data sample is that intensity lines for individual countries generally do not cross. The line representing France is always above the one representing Italy. At first glance, this may imply some “intrinsic factors” like climate conditions, differences in economic profile (agriculture, heavy industry etc.) which may explain a certain “unbridgeable” gab between countries. However, as our figure clearly indicates, it is indeed possible that country lines cross each other. UK, starting out at an energy intensity well above Italy in 1995, has succeeded to fall consistently below the Italian level. Moreover, this is not a short term fluctuation, but rather can be safely considered a consistent trend. This, in turn, indicates that energy saving measures may have a significant impact on the efficiency of energy usage.

The decoupling of economic performance and energy consumption can be seen in the following two figures referring to Germany and Denmark, respectiveley. In order to facilitate the visibility of the effect we had to adjust the figures somewhat as will be explained immediately. Fig. 2 shows the case of Germany during the period 2001-2009.

Fig. 2 Germany´s inland consumption vs. GDP

The figures for the GDP are given in G€, whereas – for better visibility – inland consumption has been scaled as Mtoe*5. This puts the two curves close to each other and clearly indicates the respective trends.

Fig. 3 shows a similar pattern for Denmark. Here again, the GDP is plotted in G€, and inland consumption is put on a scale of Mtoe*10.

Fig. 3 Inland consumption and GDP in Denmark

Both, Fig. 2 and Fig. 3 show the impact of the economic crisis starting in 2008 on economic output and inland consumption. Nevertheless, during the years before the financial crisis it is obvious that an increase in GDP comes together with a decreasing energy consumption.

One question to be asked is whether there is a lower limit to the energy intensity which cannot be undercut. One is inclined to think that a country´s level of energy intensity may be largely determined by factors such as climatic conditions, the level of industrialization etc. However, it is possible that northern countries may “beat” the southern ones, as the case of UK and Italy indicates. Moreover, there are substantial differences even between countries situated a similar latitudes like Italy and Spain. This, in turn, may indicate that there is still a considerable potential for improvement in the case of Spain.

Summing up, we can conclude that it is possible for developed economies to have a growing GDP while at the same time keeping energy consumption stable or even lowering it.