Electricity lost along the way

Why the grid raises voltage, and where the losses end up as heat. Theory, formulas, worked examples and applications.

What is lost does not disappear: it heats the cable

Why any real cable loses energy

When electric current flows through a conductor, the moving electrons constantly collide with the atoms of the material's crystal lattice, and each of those collisions transfers a small amount of energy as heat. This phenomenon — the Joule effect — is not a manufacturing flaw fixable with a better material: it is a direct consequence of the fact that no real conductor has zero electrical resistance at room temperature. The amount of energy dissipated grows with the square of the current flowing, which has an enormous practical implication: to transport the same electrical power, halving the current cuts Joule-effect losses to a quarter.

The engineering solution: raise voltage, lower current

Since electrical power is, essentially, the product of voltage and current, there is a direct way to transport the same amount of power with much less current: raise the voltage. This is exactly what the grid does on leaving a power plant: a transformer raises generation voltage to levels of hundreds of thousands of volts for long-distance transport, precisely to minimize losses along the way, and steps it back down in several successive stages — at transmission substations, then distribution substations — before electricity reaches a voltage level safe for household or industrial use.

What factors make a cable lose more or less

The amount of energy dissipated in a specific line depends on several combined factors: the conductor material (copper and aluminum have different resistivities, and aluminum, lighter and cheaper, dominates long-distance transport despite conducting somewhat worse than copper, compensating with a larger diameter), the line's total length, ambient temperature and the cable's own temperature from carrying current (a hotter cable has more resistance, which in turn slightly increases losses further), and how much load flows through it at any given moment. Transformers, in addition, have their own losses: fixed losses that exist even with no load connected — from the core's own magnetization — and additional losses that grow with the load passing through them, just like a cable.

Saturated grids lose more and age faster

A line or transformer that habitually operates close to its maximum design capacity not only loses more energy proportionally, but also ages faster: the extra heat generated accelerates the wear of insulation and other components, shortening the equipment's useful life. This is why grid operators design a line's capacity not only for current peak demand, but also with a margin for future growth and the need to maintain safe operating temperatures long-term. Bringing generation closer to the point of consumption can reduce transport losses, but a wide, well-interconnected grid allows resources to be shared between regions and reduces the amount of dedicated backup capacity each area would need to build separately — a trade-off between local losses and system-wide resilience.

Why not all possible loss is eliminated

Reducing a specific line's losses is almost always technically possible — a thicker conductor, with more copper or aluminum cross-section, loses less energy per unit length — but it is not free: it uses more metal, costs more to manufacture and install, and that extra upfront cost only pays off over time, through the energy savings accumulated over years or decades of operation. Electrical grid engineering largely consists of this constant exercise of weighing today's material and construction cost against tomorrow's energy savings, along with the transport capacity needed and the demand growth expected over coming decades.

Household savings have an invisible tail

When someone saves one kilowatt-hour at their own socket, they avoid generating quite a bit more than one kilowatt-hour at the source plant, precisely because neither the transport grid nor the intermediate transformers are perfect: part of what was generated would never have reached the socket anyway. The exact size of that extra margin changes by specific grid and moment, and there is no point memorizing a universal percentage — but the general lesson always holds: efficiency in final consumption and efficiency of the grid itself reinforce each other, they are not two independent things.

A thicker cable is not always the right answer

In favor of investing in lower losses: a larger cross-section conductor loses less energy every year over decades of operation; on very long or very high-utilization lines, that accumulated saving can far outweigh the extra material cost.

To watch: every improvement in cross-section or voltage has an immediate material and construction cost — more copper or aluminum, sturdier towers, higher-capacity transformers — that only pays off if the line is actually used with the intensity that justified the investment.

The mistake of comparing countries by a single loss percentage: a grid with higher transmission and distribution losses is not necessarily worse managed; it may reflect greater geographic distances between generation and consumption, an older grid still being modernized, or a more dispersed demand pattern. Comparing that percentage without that context leads to hasty conclusions.

Before proposing "distributed generation" as a universal fix for losses: bringing generation closer to consumption does reduce transport losses for that specific energy, but a wide, well-interconnected grid provides resilience and the ability to share resources between regions in return — a trade-off that does not always favor purely local generation.

The numbers, with their source

About one in every twelve units of electricity generated

Worldwide, transmission and distribution losses average around 8% of all electricity generated — meaning roughly one out of every twelve units of energy produced at a plant never reaches its destination, dissipated as heat along the way. That figure varies enormously by country: the most efficient grids, with fewer technical losses and less electricity theft, keep losses between 5% and 8%, while some developing-country grids, with older or more dispersed infrastructure, can exceed 18%.

Source: World Bank / IEA — Electric power transmission and distribution losses (% of output).

Why this figure is not only a technical problem

Part of these losses is unavoidable by pure physics — no conductor has zero resistance —, but another part reflects investment decisions: older grids, with conductors thinner than ideal for current demand, obsolete transformers, or habitually overloaded lines, systematically lose more energy than a modernized grid. In some countries, part of the official "losses" figure also includes electricity consumed without payment — illegal connections or metering fraud — mixed statistically with purely physical loss, which complicates comparing countries by this number alone.

The table connecting this figure to the rest of the section

What this table is missing

The global 8% figure mixes transmission and distribution into a single number; separating them country by country requires more detailed data that grid operators do not always publish, making it harder to pinpoint exactly which stretch of the grid is worth investing in first to cut losses.

Sources

• The US Energy Information Administration explains electricity grid losses.
• The World Bank maintains the international indicator for electric power transmission and distribution losses by country, sourced from the IEA.
• ENTSO-E documents the design and planning of European transmission infrastructure.