Why NASA is struggling to get its most powerful space telescope off the ground

For more than two decades, NASA has been developing what is being hailed as the most powerful space telescope ever created, a technological masterpiece that will live 1 million miles from Earth and unlock the mysteries of the distant Universe. The James Webb Space Telescope (JWST) will have a 25-foot golden mirror that will be able to collect light from the first stars and galaxies that sprung to life just after the Big Bang. But the space agency just can’t seem to get the telescope off the ground.

Since 2011, NASA had hoped to launch the JWST sometime this October. But in June, the space agency announced that the project would not launch until March 2021, and it will require millions of dollars more than NASA currently has budgeted.

It’s another nasty sting to the program, which has become defined by its history of schedule delays and cost overruns. JWST was originally conceived in 1996 as a $1 billion telescope, with an expected launch sometime around 2007. Since then, the scope of the project has ballooned: its total lifetime cost will be more than nine times that much, around $9.66 billion, with a launch more than a decade later than planned. Now, many are looking for someone to blame for the project’s woes.

In recent months, much of the focus has centered around the telescope’s primary contractor Northrop Grumman. The main components of the JWST are currently being assembled at the company’s facilities in Southern California. But numerous humans errors at Northrop Grumman over the last few months have stalled the telescope’s development: an incorrect solvent was used to clean a fuel valve, while someone applied the wrong voltage during a test. In a recent hearing, members of Congress came down hard on Northrop Grumman, arguing that the contractor should possibly front the cash for its mistakes. “This is about the biggest screw job I’ve ever seen, and the taxpayers are getting screwed here,” Rep. Dana Rohrabacher (R-CA) said to Northrop Grumman’s CEO at the hearing.

Northrop Grumman is an easy scapegoat, but many experts argue that the blame for JWST’s woes started in the planning phase. Because JWST is so unique and complex, NASA never really had realistic estimates from the start about how much it was going to cost or how long it was going to take to build. “There was no one on the planet who knew how to build JWST when they started. That information didn’t exist,” Jim Muncy, founder of PoliSpace, a space policy consulting agency, tells The Verge. “They tried coming up with a good faith estimate for cost, but you really don’t know how much it’s going to cost because you haven’t done it before.”

The design for JWST was first proposed to NASA by the National Academies of Sciences, Engineering, and Medicine in 2001. The concept, then known as the Next Generation Space Telescope, was considered the top priority in the academies’ decadal survey, a review curated every 10 years detailing the projects that the science community would most like the government to fund. At the time, the academies estimated a modest $1 billion price tag. “The whole process that led JWST from day one was fundamentally flawed,” Charles Miller, president of NexGen Space LLC, a space consulting firm, and a former member of the Trump administration’s NASA transition team, tells The Verge. “A bunch of scientists selected the most compelling science without any understanding or true basis to cost.”

NASA has long been plagued by critically underestimating spacecraft costs. The Hubble Space Telescope, initially projected to cost $200 million, wound up costing $1.2 billion to develop. After, its mirror required multiple fixes, which racked up billions more in expenses. A recent report from NASA’s inspector general identified a culprit of NASA’s chronic underestimation: a culture of optimism. He noted that NASA does not consider meeting cost and schedule deadlines as a measure of success and that the agency expects more money to be provided if projects go over budget. That has certainly been the case for JWST, which has continued to receive funding as its costs increase. Congress capped the development budget at $8 billion, but lawmakers are still expected to fund the telescope’s new budget overruns. The space agency is also incentivized to low-ball on price when presenting projects to Congress, making them much more appealing at first blush.

Another factor in the overrun was that not all of the technology needed for JWST had been created at the time of its conception. NASA engineers had to invent up to 10 technologies before construction of the telescope could begin. These included software to keep the telescope properly positioned in space and the material for the vehicle’s sunshield, which keeps the spacecraft from getting overheated. “Anytime you invent new technologies, it’s going to be much more costly than anyone ever imagined,” says Miller. “The whole process was backwards. You should invent the technologies first and then invent the telescope. But they decided, ‘We’re going to do this. Now let’s go invent the 10 technologies.’ There was huge uncertainty when they started that process.”

Fast-forward to today, and the consequence of these underestimations makes each new delay even more painful. Because Northrop Grumman now employs a dedicated team of individuals who work exclusively on JWST, each little mistake means massive labor costs — just a few months of delay cost millions of dollars. “The program is enormously complex,” Grant Tremblay, an astrophysicist at the Harvard-Smithsonian Center for Astrophysics, tells The Verge. “It has a lot of inherent built-in risk and more opportunities for delay. And then when you get a delay, those delays are longer and more expensive. They can create a cascade of further delays.”

Northrop Grumman currently enjoys what’s known as a “cost-plus” contract with NASA. That means the contractor will be reimbursed by the government for everything that is required to build this telescope — from the personnel needed to build and test the spacecraft to the facilities and hardware that need to be created to piece everything together. It also means if you run over budget, the government will pick up the expenses.

It’s a type of contracting method that many have criticized for making programs more expensive since contractors don’t seem to have major disincentives for making mistakes. Congress has wondered if perhaps another contracting method would have been better or if Northrop Grumman should pay for the cost overruns itself. It’s this proposition that Northrop Grumman isn’t particularly keen on. “Our view on that is that would create more of a fixed-price relationship on this program, which would significantly impede and impair the relationship between NASA and Northrop Grumman,” Northrop Grumman’s CEO Wes Bush said.

NASA has started implementing fixed-price contracting lately for things like sending crew and cargo to the International Space Station. Instead of reimbursing contractors for all of their costs, NASA buys a service from a company, investing a fixed amount of money into the program’s development. The company is responsible for spending its own money on any cost overruns. Experts say such a way of doing business would not have worked for the construction of JWST, though. Fixed-price is usually best for companies that are developing a system they can use to make a profit later on. “An instrument like JWST is a one-time thing,” Laura Forczyk, a space consultant and owner of space research and consulting firm Astralytical, tells The Verge. “If Northrop Grumman didn’t have that cost-plus contract, they could not independently make money from it later on.”

Without a cost-plus contract, embarking on the development of a brand-new technology like JWST would have been a huge gamble for Northrop Grumman. “Sometimes the government is doing something absolutely new they’ve never done before. And when a contractor is doing something new, and they literally don’t know how to do it, it does make sense to use a cost-plus contract so that you figure it out together as you go along,” says Muncy.

Still, Northrop Grumman may suffer repercussions from its cost overruns. The company has said it may consider forfeiting its profits on JWST. Through cost-plus contracts, contractors are given various award fees based on how well the mission does. Northrop Grumman’s CEO said he would consider forfeiting the award fees if the mission turned out to be unsuccessful.

Of course, much of this could have been avoided if JWST had a solid estimate from the start, one that took into account potential human failures. Now, JWST’s budget overruns are putting the future of other big-budget missions in jeopardy. For instance, NASA is working on the development of another big space telescope, WFIRST, which is already running over budget. The Trump administration has proposed canceling WFIRST due to costs, and getting funding for the project may be more difficult now, as JWST will continue to dominate NASA’s astrophysics budget for at least a few more years. Meanwhile, NASA is also trying to limit the costs of other future telescope projects to avoid what has happened with JWST.

Observers note that while JWST’s ballooning budget is a concern, it’s important to maintain perspective: NASA’s budget is relatively small. The agency receives roughly half of 1 percent of the annual federal budget, around $19 billion, and astrophysics receives just over $1 billion of that. NASA isn’t the only government agency to experience cost overruns, either. The Department of Defense, for example, has shelled out at least $100 billion more than it projected for the development of a single plane, the F-35 fighter. Ultimately, JWST may be a small price to pay to see the birth of the Universe. “JWST will witness the birth of stars until the edge of time for what US consumers spend annually on potato chips,” says Tremblay.