GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Guest Blogs

Why Doesn’t the U.S. Bury Its Power Lines?

The idea is appealing, but putting utilities underground is expensive and carries its own risks in a storm

Above ground power lines are vulnerable to storm damage, but placing utilities underground doesn't solve all the problems. (Photo: Dive Officer / CC / Flickr)

It is nearing the end of a highly destructive hurricane season in the United States. The devastation of Hurricane Florence in North and South Carolina caused more than 1.4 million customers to lose power and Hurricane Michael has cut service to an estimated 900,000 customers in Florida, Alabama, Georgia, and the Carolinas. Soon, winter storms will bring wind and snow to much of the country.

Anxious people everywhere worry about the impact these storms might have on their safety, comfort and convenience. Will they disrupt my commute to work? My children’s ride to school? My electricity service?

When it comes to electricity, people turn their attention to the power lines overhead and wonder if their electricity service might be more secure if those lines were buried underground. But having studied this question for utilities and regulators, I can say the answer is not that straightforward. Burying power lines, also called undergrounding, is expensive, requires the involvement of many stakeholders and might not solve the problem at all.

Where should ratepayer money go?

Electric utilities do not provide service for free, as everyone who opens their utility bill every month can attest. All of the costs of providing service are ultimately paid by the utility’s customers, so it is critical that every dollar spent on that service provides good value for those customers. Utility regulators in every state have the responsibility to ensure that utilities provide safe and reliable service at just and reasonable rates.

But what are customers willing to pay for ensuring reliability and mitigating risk? That’s complicated. Consider consumer choices in automobile insurance. Some consumers choose maximum insurance coverage through a zero deductible. Others blanch at the higher premiums that zero deductibles bring and choose a higher deductible at lower premium cost.

To provide insurance for electricity service, regulators and utilities must aggregate the preferences of individual customers into a single standard for the grid. It’s a difficult task that requires a collaborative effort.

A model effort in Florida

The state of Florida’s reaction in the wake of the 2004-2005 hurricane seasons provides a model for this type of cooperative effort. Utilities, regulators, and government officials meet every year to address the efficacy of Florida’s storm hardening efforts and discuss how these efforts should evolve, including the selective undergrounding of power lines.

This collaborative effort has resulted in the refinement of utility “vegetation management practices” — selective pruning of trees and bushes to avoid contact with power lines and transformers — in the state as well as a simulation model to assess the economic costs and benefits of undergrounding power lines.

Nationally, roughly 25% of new distribution and transmission lines are built underground, according to a 2012 industry study. Some European countries, including the Netherlands and Germany, have made significant commitments to undergrounding.

Burying power lines costs roughly $1 million per mile, but the geography or population density of the service area can halve this cost or triple it. In the wake of a statewide ice storm in December 2002, the North Carolina Utilities Commission and the electric utilities explored the feasibility of burying the state’s distribution lines underground and concluded that the project would take 25 years to complete and increase electricity rates by 125%. The project was never begun, as the price increase was not seen as reasonable for consumers.

Construction at the Moody Air Force Base in Georgia to put power lines underground in 2009. (Photo: U.S. Air Force, Senior Airman Schelli Jones / CC)

A 2010 engineering study for the Public Service Commission on undergrounding a portion of the electricity system in the District of Columbia found that costs increased rapidly as utilities try to underground more of their service territory. The study concluded that a strategic $1.1 billion (in 2006 dollars) investment would improve the reliability for 65% of the customers in the utility’s service territory, but an additional $4.7 billion would be required to improve service for the remaining 35% of customers in outlying areas.

So, over 80% of the costs for the project would be required to benefit a little more than one third of the customers. The Mayor’s Power Line Undergrounding Task Force ultimately recommended a $1 billion hardening project that would increase customer bills by 3.23% on average after seven years.

Shifting risk

In addition to the capital cost, undergrounding may make routine maintenance of the system more difficult, and thus more expensive, because of reduced accessibility to power lines. This may also make it more difficult to repair the system when outages do occur, prolonging the duration of each outage. Utility regulators and distribution utilities must weigh this cost against the costs of repairing and maintaining the electricity system in its overhead state.

Electricity service is valuable. A 2009 study from the Lawrence Berkeley National Laboratory estimated an economic cost of $10.60 for an eight-hour interruption in electricity service to the average residential customer. For an average small commercial or industrial customer the cost grew to $5,195, and to almost $70,000 for an average medium to large commercial or industrial customer. The economic benefits of storm hardening, therefore, are significant.

Beyond the economic value of undergrounding, one could consider other benefits, such as aesthetic ones, which may be more difficult to quantify. The safety of the electricity grid is also a concern. The California Department of Forestry and Fire Protection recently concluded that high winds and above-ground power lines were the cause of the Cascade Fire of October 2017. But all costs and benefits must be considered to ensure value for the customer’s investment.

In terms of reliability, it is not correct to say that burying power lines protects them from storm damage. It simply shifts the risk of damage from one type of storm effect to another.

For example, it is true that undergrounding can mitigate damage from wind events such as flying debris, falling trees and limbs, and collected ice and snow. But alternatives, such as proper vegetation management practices, replacing wood poles with steel, concrete or composite ones, or reinforcing utility poles with guy wires, may be nearly as effective in mitigating storm damage and may cost less.

Also, undergrounding power lines may make them more susceptible to damage from corrosive storm surge and flooding from rainfall or melting ice and snow. Areas with greater vulnerability to storm surge and flooding will confront systems that are less reliable (and at greater cost) as a result of undergrounding.

So, the relocation of some power lines underground may provide a cost-effective strategy to mitigate the risk of damage to elements of a utility’s infrastructure. But these cases should be evaluated individually by the local distribution utility and its regulator. Otherwise consumers will end up spending more for their electricity service, and getting less.


Theodore J. Kury is the director of energy studies at the University of Florida. This post originally appeared at The Conversation.


  1. flyingbytheseat | | #1

    Interesting! Our lines are underground where we are building our new home and it makes sense for that area (heavily wooded, significant hills/small mountains). I do appreciate the aesthetics of not seeing power lines everywhere too. The expense is considerable for homeowners and builders too since the line has to be buried to the house and to any other structures that need power, it's an investment all around

  2. AntonioO | | #2

    Somewhat thin on storm risks for an article subtitled: The idea is appealing, but putting utilities underground is expensive and carries its own risks in a storm

    In my neighborhood, for a long time power would go out if anyone sneezed too intensely. Recently, the power lines were underground-ed to use the language of the author. It made a huge difference. The lines were placed in conduit (similar in appearance to the material used in a neighbor's GSHP tubing) that was bored--no open trenches required for the most part. I believe the plastic conduit will go a long way towards avoiding the corrosion mentioned in the article and leave the lines less susceptible to flooding concerns. Also, the power lines were placed after the conduit, so it appears to be possible to replace a line without digging up the conduit. So, as I see it the big drawback is cost; I imagine that repairing poles (AND LINES) after a storm comes at considerable costs as well. In addition, cutting back trees every couple years, as in my neighborhood cannot be inexpensive.

    1. Expert Member
      BILL WICHERS | | #4

      Since I work in a very closely related field (telecommunications) I can speak to this a little. The conduit fills with water. Always. Nothing you can do to stop it. Look in just about any manhole and you’ll see water. The conduit links all the manhoods together so the water spreads throughout the system. The corrosion issues are at splices, terminations, basically everywhere connections are made. It’s difficult to seal everything completely. Underground lines essentially have different problems from aerial lines, not less problems. The conduits also can become crushed or otherwise obstructed in which case repair is vastly more difficult. Imagine a crushed duct bank (usually the conduit runs are in large groups of many runs along a route) under a bush street. How do you fix it? Well, you have to close at least part of that busy street, sawcut the pavement, excavate the duct bank, correct the problem, put everything back together, repave the street. This can take a week or more. An aerial line across a steet can be repaired in a manner of hours.

      Underground lines can also have more loss (be less efficient) due to inductive and capacitive effects that are much less pronounced in aerial lines. This can result in limited maximum line length (this is why HVDC, and not AC, is required for undersea power cables of more than a few tens of miles).

      Like most anything else engineering related, Undergrounding lines is a much more complex issue than most people outside the industry realize. Directional drilling, mentioned above as a nice way to avoid trenching, is expensive and occassionally hits things. My own crews have hit high pressure gas lines in the past, among other things. The technology is great, but not without risk.


      1. AntonioO | | #7

        That's interesting. Where I live, power and telecommunication lines generally run along side the streets (when underground), not under them--except for crossings at intersections--meaning that the probability of a problem under a street is statistically much, much less.

        I also wish you could expand on the inductive and capacitive losses being more pronounced in underground lines. Why is that? If the conductors are run separately (assuming you are really talking about high voltage, long distance lines here) and sufficiently spaced, I don't see how underground [long distance] transmission would be inferior.

        As for hitting things, trenching hits things as well. By the way, the argument is not that undergrounding is not without complexity or risk. The argument is that the complexity does not appear greater than hanging power lines on poles and leaving them susceptible to frequent weather related events and other above ground hazards. If complexity were really the main problem, above ground lines in new development would be more prevalent. The problem is the cost of "upgrading" existing above ground distribution lines to underground distribution. Finally, I acknowledge a difference in the examination applied to long distance transmission as opposed to neighborhood distribution lines.

        1. Expert Member
          BILL WICHERS | | #10

          It’s always preferable to run the lines in the “green belt” on the side of the road, but that’s not always possible. Think of the case in an urban environment where everything is paved (ugh!). The other issue that’s much more common now than a few decades ago is that the underground rights of way are getting crowded. Sometimes there just isn’t enough room for whatever new cable needs to be installed so it goes out under the street. In the case of power systems, many duct banks are 50 or more years old. If the road has been widened, the duct bank that was once in the green belt is now under the road. This is actually very common. BTW, as a design engineer working with these systems, I can also tell you that accidental damage is much less common for systems installed under the road. Gardeners and homeowners who don’t call miss dig don’t dig in the road :-)

          I can expand on the inductance and capacitance. I’m an EE. It gets complicated though. High voltage cables for use underground are generally of a concentric design, similar to coaxial cable like cable TV uses. The conductor is covered with a semiconductive layer, then a thick layer of insulation, another semiconductive layer, a spiral shield consisting of multiple wires, then the outer jacket. The shield is to prevent the outer surface of the cable from becoming electrically charged. The semiconductive layers are to even out the electric field (think of it as preventing any pointy parts from sparking) to limit stress in the thick insulation layer. Stress can result in “punch through” which is how the cables fail. A typical 2 gauge concentric cable rated for 15,000 volts is about 1” in diameter.

          Aerial wire is just wire. Usually uninsulated, but sometimes with a thin insulating layer to limit atmospheric corrosion. The porcelain or polymer insulators on the poles are what do all the actual insulating work for the high voltage.

          Now for the exciting part! The widely spaced aerial wires have very low capacitance between them due to both the spacing and the fact that air is a good dielectric (fancy way to say insulator). The concentric cable has a FAR higher capacitance. The inductance is a function of the diameter of the wire, for the most part. There is what is known as “impedance” which is a phenomonon of AC carrying wiring that is a combination of the effects of the inductance and capacitance of the wiring. Losses come from both resistance of the wire AND dielectric losses from the insulation. Dielectric losses are because the “capacitor” formed by the insulation isn’t perfect. More energy lost as heat.

          AC power in North America operates at 60 cycles per second, each cycle has to charge up the “capacitor” formed by the wiring, then discharge on the second half of the cycle (wave crest actually), but some power is lost each time. In large power system design, some of the underground cabling capacitance is actually used to accomplish power factor correction on parts of the power grid. AC systems are more complex than DC systems in this case.

          If you want to read more, since I’m not sure how long a post here can be, lookup “HVDC” power transmission (HVDC is high voltage DC). ABB is a large manufacturer of this stuff and has some good info on how it works and also about some of the limitations of AC cabling in certain situations. You may find it interesting reading if you’re curious.


  3. Jon_R | | #3

    For less than a 125% increase in electric bills, I can buy/install/maintain/fuel a generator for use during outages. With less risk (buried lines still have outages).

    > economic cost of $10.60 for an eight-hour interruption

    As mentioned, the actual value is much more complicated than this.

    Not clear to me is the role and cost of redundant distribution lines (eg loop layouts down to the neighborhood level).

    1. Expert Member
      BILL WICHERS | | #5

      Many utilities have been installing redundant lines, and other means to reduce failures. The devices used are known as “sectionalizers” and are used to essentially split lines into many smaller areas that can be individually turned on and off to isolate outages to smaller numbers of people. This is a happy medium cost wise to improve reliability of the system for relatively low cost. True looped (ringed redundant circuits) are relatively rare and are usually only used for very large customers, customers with critical loads that pay for the service (I’ve seen quotes of over $700,000 to install such service), or large infrastructure like substations that support large numbers of customers.

      Everything has to get a cost/benefit analysis. It’s usually cheaper for customers to just get small generators. I personally would rather put in a generator for essentially a one-time cost and some minimal ongoing maintenance costs and not pay much higher utility rates forever. Outages are not frequent enough to justify many of the costs to improve reliability.


      1. Jon_R | | #6

        I'm guessing that in a case like my neighborhood (a big oval with lots of outage causing trees), the extra cost comes from up-sizing the lines to be able to handle the entire neighborhood vs two far-end-unconnected lines, each handling 1/2 of it.

        1. Expert Member
          BILL WICHERS | | #9

          Not really. The wire itself is only a small portion of the total project cost, and at the high voltages running on the primary (usually 7,200 volts or more in modern systems) make the current requirement per house relatively small. The industry uses an average of 1kw per house, let’s be conservative and say you’re in a really upscale area and use 2kw per house. Assume you’re in a development with 100 houses. That’s 200kw of avarage load. That’s about 28 amps of load on a 7,200 volt circuit. The smallest common underground cable for the primary (high voltage) is 2 gauge and is rated for a bit over 100 amps. The smallest common aerial conductor is 6 gauge and is good for almost the same amout since it’s in free air and can cool off better.

          Basically the physical strength of the aerial conductor is what’s important here, any size that would be used will have plenty of current carrying capability.


          1. Jon_R | | #14

            Evidently that leaves no good reason that my neighborhood that naturally lends itself to a loop wasn't wired as a "ringed redundant circuit".

  4. jackofalltrades777 | | #8

    My rural subdivision is all underground utilities. So on my house build I had to pull power from a power box at the base of my lot and up my driveway for about a 1/2 mile to my house site (large lot - 15 acres). I was responsible for trenching, conduit, fill and covering. APS would only cover the cost of the actual wire and transformer. APS also did the labor to push the wire through the conduit and connect power.

    Once you lay down & glue the conduit in the trench. APS shows up and used compressed air to blow a foam "missile" through the conduit line to make sure it's clear of obstructions. If it tests good, they then fish the electrical cable through the conduit. Every 600 feet they install a grade level pull-box.

    Trench had to be around 4 feet deep with around 18 inches of fines on top of the conduit before back-filling. Bids were around $15k-$20k or more to do the work so I ended up doing it myself. Rented an excavator from United Rentals, got the conduit at Lowe's, used fill from a wash (sand and small smooth rocks) and went at it. Cost me around $5k in materials and rentals. Saved myself $10k-$15k

    It's so much more aesthetically pleasing with it underground. Only time the power went out was when the ABOVE GROUND lines miles away were struck by high winds.

  5. AlanB4 | | #11

    This is a very interesting article. I assumed that underground was superior in most ways but was just more expensive. Learning its shortcomings is quite interesting.
    That said most would probably think the same as i did because its a rather intuitive conclusion.

    But the huge cost difference and the other comments really point to above ground being superior overall.

    That said Bill's idea about a generator or if other blackout measures were in place (say energy storage batteries) it would be more cost effective and probably even more resilient then underground lines.

    1. Expert Member
      BILL WICHERS | | #12

      I would say, to summarize things in a more GBA common lingo, is R250 insulation in your attic going to help keep your house warm? Yes. Is it cost effective over R49? Not so much. You’d be better off using the savings on something else.

      My own opinion is the underground lines are best for entering structures (for a number of reasons), and for keeping a clean appearance. Reliability wise there are other issues. A friend of mine in the telecom industry, who started in 1967, said that in his expierience aerial and underground weren’t much different in terms of reliability. Aerial is more prone to weather damage, underground is more prone to accidential damage.

      Generators are best since they protect against failures in wire and other devices.


      1. AlanB4 | | #13

        I like that insulation analogy :D

        I like clean but i won't pay millions for aesthetics when there is so much more that money could do better. I don't care that much about underground accidental damage because line induced power failures are the biggest problems during emergencies, weather events, trees falling, Ice/snow etc. If a natural gas company broke my underground power line then i'd end up at a hotel for a day or a week which they would have to pay for. Inconvenient but in the end small potatoes

  6. kjmass1 | | #15

    This article is timely as we lost power with Saturday’s storm from another fallen tree- 3 times in one year. We have 80+ year old oaks every 40’ or so on our street, and their root structure has started to give way against the concrete curbing and streets.

    That said, the tree took down a transformer, pole, and all wires. We had temporary power back within 8 hours and internet in 16 hours. All new pole installed and rewired within 2 days. I’m sure that cost is a fraction of what it would be to trench up the street and run dedicated lines to each house in a very dense neighborhood of 50’ wide lots.

    Definitely starting to look in to back up generators.

  7. jackofalltrades777 | | #16

    Let's not forget above ground wires keep the birds & squirrels happy and the power worker employees, busy & employed. I still prefer underground, both for aesthetics and storm resistance.

    1. AlanB4 | | #17

      Underground would probably keep them better employed, adjustments or repairs are often not a simple matter.
      I would like underground if its cheaper and more resilient but its a tradeoff, how much are you willing to pay for some extra resilience. Its not absolute.

  8. Expert Member
    MALCOLM TAYLOR | | #18

    If you could time-travel, it would be fun to try and describe electrical power transmission to people in the 19th century.

    We have found an invisible power source for industry and homes. We propose supplying it to every building in the country by connecting them via copper wires. To keep the wires out of the way, we will suspend them on tall sticks and run these along roadways.

    The look on their faces would be priceless.

    1. AlanB4 | | #22

      Don't tell them about cell phones or they might think your a tool of the devil and that never ended well.

  9. jackofalltrades777 | | #19

    While it was not discussed here. High tension power lines produce strong magnetic fields. Nobody wants to live near these things and property values drop when a home is located near them. Some studies show certain cancers are linked to long term exposure. I've seen some expensive homes be built near these towers. No thanks. I'll pass. The health concern is a major thing but aesthetically looking at those towers out my window is not appealing to me.

    1. Expert Member
      Michael Maines | | #20

      Peter, I agree, but have also heard that the cancer rates for those who live near high-tension lines may have something to do with the fact that the area below the lines is always magically clear of broad-leafed plants, due to liberal application of herbicides like Roundup.

    2. Expert Member
      BILL WICHERS | | #21

      I don’t buy the health issues with these lines. There are third generation workers in the electric industry that have spent their entire working lives around the lines and sometimes also in the plants near multi-hundred megawatt generators. Much of the magnetic field is cancelled by both the transmission line characteristics (which has to do with impedance, not just that the powerline is called a “transmission line”), and also the opposing phases in the three phase system. There is very, very little radiated energy.

      I agree with the aesthetic issue. I wouldn’t want to live right next to these lines myself. Undergrounding the very high voltage transmission lines (usually 120kv and up, the highest standard voltage currently in use in North America is 765kv) is a MUCH bigger deal than Undergrounding distribution lines (around 7.2 - 13.8kv). The underground cable for the very high voltages is massive due to all the insulation, around 6+ inch diameter PER WIRE.


  10. kyeser | | #23

    I am having a hard time believing it costs a 1/2 million to a million dollars a mile to bury lines. An excavator can dig a long way in a day. Conduit is cheap and it doesn't take that long for to pull wire either.

    My municipality has figured out that it is cheaper in the long run to maintain a system that is buried. So all new work is buried. Less power outages, less to fix after wind/ice storms.

    I'll dig all the power lines you want for $1,000,000 a mile. Somebody's making a killing.

    1. jackofalltrades777 | | #24

      The real cost to do it probably $250,000 a mile but 2 to 4 times multipliers is the upcharge. I was quoted a 3 to 4 times multiplier to do the underground power pull on my lot. Cost me $5k to do it myself. Quotes were coming in at $15k - $20k

    2. Expert Member
      BILL WICHERS | | #26

      The cost depends on many factors. Can it be installed by open trenching? Many cities require the use of directional drilling now since it’s less disruptive — but it costs more. At work, where i contract this work out, we pay about $12-15 per foot (about $63,000-$79,000 per mile) to install 3 x 1.25” conduits. We put in fiber optic cable, the number I give is for installation of the conduit only. Power companies will install 4” or 6” conduits, and theirs have to be heavier walled with a different rating. Small precast manholes are about $1,000 each. Power companies need big ones.

      Sawcutting roadways for crossings is expensive. So is repaving. Many road commissions won’t allow sawcutting and require us to replace full sections (10x20 feet) of concrete. These are 8” thick slabs. We also have to pay for traffic management, flagmen, everything else. This adds up.

      Fiber optic cable is cheap. High voltage concentric power cable is not. This is not wire like used for home services, it’s specialty stuff (see for some examples). Reel lengths are limited, so you have to have sealed splices in the manholes. More stuff to buy, more labor costs.

      Aerial cable is aluminum and steel, usually no insulation (a common type used for high voltage lines is ACSR, aluminum conductor steel reinforced). Even with poles and insulators, it’s far cheaper than underground.

      I’ve seen fiber optic installation quotes of around a half million per mile in city centers (its very congested under the roadways in city centers so it’s more expensive to do the work). Power cable installation costs more.

      For distribution lines (4.8 - 13.8 kV) in a suburban area, I’d guess around $250,000 per mile is probably close. For transmission lines (120+ kV), $1 million per mile may well not be enough.


  11. GBA Editor
    Martin Holladay | | #25

    The cost varies depending on whether you are talking about an uninhabited stretch of rural road, or a densely inhabited suburb. If there is a house or commercial establishment every 100 feet, you'll need to consider the cost of new transformers and new connections to each house.

    1. Expert Member
      MALCOLM TAYLOR | | #27

      "the cost of new transformers"
      And where they will go. Areas with underground services typically have transformers mounted on pads near the street.. Especially in urban areas, finding space for them could pose a problem.

      1. Expert Member
        Dana Dorsett | | #28

        In many high density urban areas the transformers are in open vaults below sidewalk level, with a ventilation grilles/grades at street level for convective air cooling. Next time you walk over a sidewalk grille in a city observe closely (visual, auditory, & olfactory) and you might be able to make out if it's a transformer vault. Entire mini-substations sometimes are place below the pavement of urban streets.

        A couple of decades ago I read a story about a transformer failure in one of those vaults in NYC caused by overheating. Apparently the people at the abutting restaurant had been dumping used cooking oil in what they thought was a drain (rather than pay to have it carted away and disposed of properly/legally.) Up to a point the thermal conductivity and convection of the oil was helping keeping the transformer but when it covered most of the cooling fins it eventually went the other way. The restaurant owners had a pretty serious repair bill and legal fines to pay- I never got the final update on it. I imagine the clean up and replacement was a fairly messy job.

        The trend these days is ever more toward "non-wires alternatives" to transmission grid capacity and resilience issue, putting distributed (smaller) generation and storage out on the distribution grid and moving away from the traditional hub & spoke approach to power generation and distribution. Self-islanding control systems and distributed storage to handle grid disruptions in frequency & voltage perturbation as transmission & distribution infrastructure fails can do quite a bit for keeping at least some or even most of the lights on as things get repaired.

        1. Expert Member
          MALCOLM TAYLOR | | #29

          In Vancouver several dogs have been electrocuted walking on those grills. They did a survey of the downtown area and found over a hundred spots where there were live grills and manholes.

          1. Expert Member
            Dana Dorsett | | #30

            Unfortunately that problem isn't unique to Vancouver, and it's not only canines who fall victim.



  12. Expert Member
    MALCOLM TAYLOR | | #31

    Would make a good plot for a mystery novel.

    1. timslager | | #32

      We lived in Germany last year and had some terrific storms, but never lost power. This article is pretty vague about what bad weather-related problems "may" occur.

      1. Expert Member
        Deleted | | #33


Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |