How Serious Is the Universe’s Expansion Problem?

With consistent, impressive leaps in observational technology, cosmology keeps running into all kinds of issues. One infamous problem is called the Hubble tension. Namely, the two main ways scientists measure the Hubble constant, which represents the universe’s expansion rate, don’t agree. Whether that’s a scientific skill issue or evidence of some unidentified physics, scientists also don’t agree.

To be clear, the numerical difference is not that big. The method that uses the cosmic microwave background (CMB), or the leftover radiation from the Big Bang, has the constant at 41 or 42 miles (67 or 68 kilometers) per second per megaparsec (a unit of distance about 3.3 million light-years). The other approach that uses local observations of galaxies and supernova puts it at 45 miles (73 kilometers).

But it’s certainly “far larger than can be explained by statistical uncertainty,” as NOIRLab notes in a recent statement on measuring the constant. So for this Giz Asks, we asked the experts for their takes on the Hubble tension. How has the academic debate over the tension progressed in recent times? What would it take to solve this problem, if that’s even feasible? Most importantly, how important is the tension to cosmology as a whole—and what’s really at stake?

The following responses may have been lightly edited and condensed for clarity.

Caroline Huang

Astrophysicist, Center for Astrophysics | Harvard & Smithsonian.

The Hubble constant is defined as how fast the universe is expanding at the current day. Right now, when we compare the Hubble constant measured through direct, local observations with the result that we derive by combining early-universe observations with our standard cosmological model, we find that these two values don’t agree.

This has definitely grown into a significant problem over the past decade. Despite all of our efforts, it doesn’t yet look like we have a satisfying modification of our cosmological model that brings the early and late observations into alignment. Many current theories require fine-tuning or create problems elsewhere. But as we showed in our recent paper, if we assume that the late universe measurements are wrong and all biased too high, we would also need to have something (or maybe many somethings) that can systematically offset over a dozen standard candle and standard ruler based probes in the same direction, which also looks increasingly daunting. None of the most robust late universe measurements scatter below the value inferred by observations of the early Universe, which you’d otherwise expect if they were both measuring the same number.

Either way, I think the resolution will be a big deal. If the model is missing something, we need new physics to augment it, but if the local measurements are wrong, it means that we have systematically misunderstood our astronomical observations across completely independent methods on a huge scale. Regardless of the outcome, solving this problem will fundamentally change how we understand the universe.

Adam Riess

Astrophysicist, Johns Hopkins University; co-winner of the 2011 Nobel Prize in Physics “for the discovery of the accelerating expansion of the Universe through observations of distant supernovae.”

The Hubble tension is one of the most significant unresolved problems in the field. For more than a decade, astronomers have tested whether the discrepancy arises from measurement errors, yet it has persisted through increasingly precise observations and independent methods. Most recently, the distance network has shown that it cannot be attributed to an error in any one tool, method, team, or telescope.

What makes the tension important is that measurements of the present-day expansion rate consistently differ from predictions based on observations of the early universe. Both approaches are precise and well-tested, but they do not agree. This could point to an unlikely confluence of subtle effects or it could be a clue that our standard cosmological model is incomplete.

Either way, the Hubble tension is doing exactly what good scientific puzzles should do: forcing us to question assumptions, improve measurements, and deepen our understanding of the universe.

Raul Jiménez

Cosmologist, University of Barcelona in Spain.

This is a very interesting question, because all current evidence points to a real tension between two ways of inferring the present expansion rate of the universe. On the one hand, local measurements based on the distance ladder find a value of the Hubble constant higher than the one inferred from the standard cosmological model, ΛCDM, when calibrated by the early universe. On the other hand, it is important to remember that almost all such measurements are, in one way or another, measurements of distance. At cosmological scales, distance is not a directly observed quantity: it is inferred within a space-time metric. Thus, what appears as a disagreement about the expansion rate may also contain hidden assumptions about geometry, calibration, astrophysical systematics, or the way we reconstruct distances across cosmic time.

This is why I think independent methods are essential. One approach that I helped pioneer, the cosmic chronometer method, does not rely on distances. Instead, it measures how cosmic time changes with redshift, providing a genuinely distance-free reconstruction of the expansion history of the universe. At present, this method does not yet have the precision to rule out the local measurements, but its results tend to agree more closely with ΛCDM. It is also true that recent large-scale-structure and CMB analyses have moved the ΛCDM-inferred value of [the Hubble constant] upward, to around 69 km/s/Mpc, closer to the local distance-ladder value. So the jury is still out. I would not yet call this a crisis, but rather a serious and fascinating problem in precision cosmology. The devil is in the details, and here the details are the systematic uncertainties. We hope to answer this question soon through our ERC Synergy Grant, RedH0t, which is designed precisely to test whether the Hubble tension is new physics or a subtle problem in the measurements.

Arthur Kosowsky

Cosmologist, University of Pittsburgh.

To measure directly the expansion rate of the universe, we need to determine distances to distant astronomical objects. That is hard! In fact, it is historically the hardest problem in astronomy. Current measurements of the Hubble parameter based on distances to various types of stars do not all quite agree, even though they are measuring the same quantity. So more hard work to identify and correct possible small systematic errors will have to continue at least until all of the different methods of measuring the Hubble constant agree.

The “Hubble tension” refers to the comparison of these astronomical determinations with a completely different way to determine the Hubble parameter. We can compare precise predictions of a very simple cosmological model with precise measurements and get outstanding agreement. I’m not losing any sleep over a possible discrepancy between the cosmological model value and an astronomical measurement value until all the astronomical measurements converge on an inconsistent value. (And maybe not even then—again, the astronomical measurements and their interpretation are fundamentally very hard in a way that the CMB measurements are not.) I’m old enough to marvel at how close the agreement is!

Niayesh Afshordi

Astrophysicist, University of Waterloo in Canada.

I have a bet on Hubble tension with my former PhD supervisor, David Spergel, on whether Hubble Tension or Black Hole echoes, will be established as evidence for new physics. Our bet goes back to 6 years ago (if I remember correctly), but the situation hasn’t really clarified since. The Hubble tension is still going strong (I believe at 6\sigma now), but it primarily relies on supernova observations, and the potential weakest link is whether they are calibrated correctly by matching to Cepheid distances measured in a relatively small set of galaxies. Other cosmological observations, however, haven’t really supported the tension, and despite many attempts, no theoretical model can satisfactorily fit the large Hubble parameter measured via supernovae with other cosmological observations. Meanwhile, we have had measurements of the dark energy equation of state by DESI, pointing to a very different type of tension, and that is where model building has now focused in recent years. To summarize, the Hubble tension is still outstanding, but the smart bet (in my view) is that we may be missing an astrophysical piece in the distance ladder, not new fundamental physics.

Stefano Casertano

Observatory scientist, Space Telescope Science Institute.

For over ten years, we have realized that there is a discrepancy, the Hubble tension, between the expansion of the Universe predictedby the leading cosmological model and what we measure around us. The tension is not small: the difference in expansion rates is about 9%, with the measurements being accurate to about 1%. Until recently, one could think that some error, some unaccounted effect, might be responsible for this tension. But as more measurements accumulate, all in agreement with each other, the evidence mounts, and the possibility of explaining away the Tension as a glitch in the measurements is now minuscule.

A recent analysis of all available measurements, the Distance Network, shows that all measurements to date are consistent and that removing any piece of the evidence, even the most significant ones, leaves the tension in place. Some teams have argued for a reduced discrepancy, often by ignoring a large fraction of the data, but even then the tension remains. So yes, it is very serious indeed.

But is the tension really a problem? Perhaps it is just telling us that there is another piece to cosmology, something that we have not considered yet. Science often operates like that: discrepancies between theory and observations of the black-body spectrum led to quantum mechanics, and the discrepancy in the precession of Mercury’s orbit led to general relativity. The leading cosmological model itself is born of an unexpected measurement of the acceleration of the universe’s expansion. We do not yet have a full interpretation of the tension, but many ideas have been suggested, and more are under study. Eventually we will learn from the tension and discover something new about the universe—and perhaps physics itself.

Marina Cortês

Cosmologist, University of Lisbon in Portugal.

As exciting as it would be for current cosmology to find hints for new physics in observations, I believe that, sadly, the Hubble tension will not grant us this opportunity. For several years I have thoroughly researched the topic, consulted with authoritative observers, and written my findings on the possibility of the Hubble tension. Nonetheless, it is well known that experiments over the last two decades have given cosmology theorists very little to work with in the form of new parameters or exotic behavior. As such, I do understand the drive of the theory community to take the Hubble tension opportunity for new physics, passed on to them by observers, and create new models to explain the tension.

If cosmology deems the Hubble tension as a globally established result before fully exhausting the possibilities of unknown systematics and pipeline errors, we face two non-negligible risks:

(1) After the observing community validates the Hubble tension and moves it along to theorists, their focus will shift to something else. This may drastically reduce the chances of identifying an error or equivalent source of the data discrepancy in the analysis pipeline. Without scrutiny from the data community, cosmologists may be conducting years of ill-advised research, possibly a decade or more.

(2) If one or two decades down the line cosmologists are shown to have made a judgment error, they may acquire a reputation of bad stewardship of global funds awarded by governments and major research councils and science foundations worldwide.

What I observe currently: prominent data analysts in cosmology express significant frustration and lack of energy at continuing to identify either systematics or implementation errors in the pipeline. In other words, they deem the data dispute a futile plight, and their time is better used on other topics. As a theorist, I thrive on the possibility of devising novel, exciting theories for the universe, of which we are all in dire need. Much as I would professionally desire an exciting period of novel theoretical discoveries, I would need to see a much larger number of senior authorities in data analysis backing up the Hubble tension. However much we bend our research efforts, the universe always has the last word.

Giz Asks is a recurring Gizmodo series in which experts answer big questions in their own words, offering a range of perspectives on the ideas, discoveries, and debates that affect our lives and shape our understanding of the world.