The Hubble constant H₀ measures the Universe's present expansion rate. Between 67 and 73 km/s/Mpc depending on the method, it crystallises the 'Hubble tension' — a major disagreement in modern cosmology.
When we observe distant galaxies, we find they all recede from us, and the more distant they are, the faster they recede. The Hubble constant quantifies this proportionality precisely. It states: for each megaparsec of distance, a galaxy recedes H₀ additional kilometres per second.
Beware the mental picture. This does not mean the Milky Way occupies a privileged centre. Every galaxy sees the same thing: each observes its neighbours receding proportionally to their distance. Classic metaphor: a balloon being inflated. Stick dots on it — from any dot, the others recede, the farthest going fastest. It is not galaxies moving through space; it is space itself stretching.
The Hubble-Lemaître law reads:
v = H₀ · d
Concrete example. The Andromeda Galaxy (M31) lies at 0.78 Mpc. At 70 km/s/Mpc it 'should' recede at ~55 km/s. In fact it approaches us at 110 km/s, because its peculiar velocity exceeds local expansion (we are gravitationally bound to it within the Local Group). Beyond ~5-10 Mpc (past the Local Group), cosmic expansion dominates and the law becomes cleanly measurable.
The historical lineage is rich. Vesto Slipher measured the first redshifts of 'spiral nebulae' as early as 1912 (he didn't yet know they were galaxies). Edwin Hubble showed in 1924 that M31 is a galaxy external to our own, via Cepheids. In 1927 Georges Lemaître published the velocity-distance relation in French and proposed an expanding Universe — a paper that went nearly unnoticed. In 1929, Hubble republished the law with more data, and it entered history under his name. H₀ has since become the king parameter of cosmology: it alone sets the Universe's age, scale and expansion rate.
The modern value of H₀ lies between 67 and 73 km/s/Mpc. The raw number looks odd because of its hybrid units — a velocity divided by a distance is the inverse of a time. Converting: 70 km/s/Mpc ≈ 2.27 × 10⁻¹⁸ s⁻¹ ≈ 1/(14 billion years).
Hence 1/H₀ ≈ age of the Universe (up to a factor depending on Ω_m and Ω_Λ). The Hubble time t_H = 1/H₀ gives 13.97 billion years for H₀ = 70 km/s/Mpc — compared to the true age measured by Planck: 13.797 billion years. Close, but not identical, confirming that expansion is not strictly linear (deceleration in the past, acceleration since ~6 billion years ago).
The Hubble distance d_H = c/H₀ ≈ 4,300 Mpc ≈ 14 billion light-years sets the cosmological distance scale.
The dimensionless parameter h = H₀ / (100 km/s/Mpc) appears frequently in papers (e.g. h ≈ 0.7). Many cosmological quantities are expressed in units of h⁻¹ Mpc to be robust against the H₀ uncertainty.
Historical values. Hubble 1929: 500 km/s/Mpc (off by a factor of 7, due to Shapley's faulty Cepheid calibrations). Baade 1952 revision: 250 km/s/Mpc. Allan Sandage 1970s: 50-55 km/s/Mpc. Gérard de Vaucouleurs 1970s: 100 km/s/Mpc — the famous 'H₀ war'. Partial resolution by the HST Key Project (Freedman et al. 2001): 72 ± 8 km/s/Mpc. Since 2013, precision has dropped below 2 km/s/Mpc — but a structural disagreement has emerged.
This is the hot debate in contemporary cosmology. Two independent measurement methods, each extremely well constrained, yield different values. This is the Hubble tension.
CMB method (indirect, early Universe). The Planck satellite measures CMB anisotropies, fits a six-parameter ΛCDM model, derives H₀. Result: H₀ = 67.4 ± 0.5 km/s/Mpc (2018). Model-dependent.
Distance ladder (direct, local Universe). Successive calibrations: Gaia parallaxes → Milky Way Cepheids → Cepheids in nearby galaxies → Type Ia supernovae in those same galaxies → more distant Ia SNe in the Hubble flow. The SH0ES collaboration (Adam Riess et al. 2022) obtains H₀ = 73.04 ± 1.04 km/s/Mpc. Model-independent.
Discrepancy = 5.6 σ. Statistically very significant. The two values cannot coincide by chance.
Three families of explanation are in circulation:
1. Unidentified systematic in one of the measurement chains. Cross-checks (JWST 2023-2024 on Cepheids and TRGB Carnegie-Chicago method) have not made the tension disappear.
2. New physics. Early Dark Energy models, new relativistic neutrino species, atypical scalar fields. None simultaneously reconciles all constraints.
3. Local fluctuation. We might sit in an underdense region ('Hubble bubble') that locally boosts apparent expansion. Plausibility limited by simulations.
The tension is one of the strongest hints of a possible crack in the ΛCDM model. The coming decade (Euclid, Rubin, Roman, DESI) should settle it.
Five major techniques, complementary.
Distance ladder. Historic method, currently led by SH0ES (Adam Riess). Gaia trigonometric parallaxes for nearby stars → Cepheid variables (Leavitt's 1912 period-luminosity relation) as secondary candles → Type Ia supernovae as final candles reaching z ≳ 0.1. Limitation: each rung introduces uncertainty.
Cosmic microwave background. Planck fits ΛCDM to the CMB acoustic peaks, derives H₀ indirectly. Most precise (0.7 %), but model-dependent.
Standard sirens (gravitational waves). Since GW170817 (neutron-star merger, August 2017, first multi-messenger standard siren), gravitational waves provide absolute distances independent of the distance ladder. Currently H₀ = 70₊¹²₋₈ km/s/Mpc — low precision but rising.
Time-delay gravitational lenses. H0LiCOW and TDCOSMO measure the time delay between multiple images of a distant quasar lensed by a foreground galaxy. 2020 result: H₀ = 73.3 ± 1.8 km/s/Mpc. Consistent with SH0ES.
Tip of the Red Giant Branch (TRGB). Carnegie-Chicago Hubble Program (Freedman et al.). Uses an alternative standard candle (TRGB) instead of Cepheids to calibrate SNe Ia. Gives H₀ = 69.8 ± 1.9 km/s/Mpc (2021), intermediate between Planck and SH0ES.
James Webb Space Telescope (JWST, launched December 25, 2021) has been checking HST Cepheid calibrations since 2023. First results (2023) confirm the local value at H₀ ≈ 73 km/s/Mpc, ruling out most photometric systematics. The tension therefore stands.
H₀ is frequently confused with several nearby quantities.
Speed of light c. The Hubble constant has the dimension of a frequency (inverse time), not a velocity. c is a fixed speed (~300,000 km/s). Very distant galaxies can recede at apparent velocities exceeding c — this is not motion through space, it is space itself stretching, and so it doesn't violate relativity.
Hubble parameter H(z). H₀ is the present value (z = 0) of a function H(z) that varies across cosmic history. In the past, H was higher (the Universe expanded faster in absolute terms). H(z) is sometimes loosely called 'expansion rate'; H₀ is its contemporary case.
Redshift. Redshift z is what we measure on a distant galaxy. It is linked to distance via the Hubble-Lemaître law, but is not H₀ itself. For small z, v ≈ c·z, so d ≈ c·z/H₀.
Dark energy. H₀ measures the present expansion rate; dark energy drives its acceleration. The two are linked but distinct. Changing dark energy changes H(z), including H₀.
Cosmological constant Λ. Λ (dimensions 1/time²) characterizes the vacuum energy density; H₀ (1/time) characterizes the expansion rate. In a pure dark-energy-dominated equilibrium, H → √(Λ/3), but we are still in a transient regime.
Because two independent methods measure slightly different things. The Planck satellite derives H₀ from the cosmic microwave background (early Universe), within the ΛCDM framework: 67.4 km/s/Mpc. The SH0ES collaboration measures H₀ locally via the distance ladder (Cepheids + Ia supernovae): 73.0 km/s/Mpc. The ~9% gap is statistically significant (5σ). Either one method contains an unidentified bias, or the ΛCDM cosmological model is incomplete. This is the famous 'Hubble tension'.
Georges Lemaître, in 1927, in a little-read French-language paper. He established the velocity-distance relation there and proposed an expanding Universe. Edwin Hubble republished the result in 1929 with more data, and it entered history under his name. Since 2018 the IAU officially recommends 'Hubble-Lemaître law' to credit both discoverers. Vesto Slipher also deserves mention: he measured the first galaxy spectral shifts as early as 1912.
Yes, beyond a certain distance. That distance (the Hubble radius) is c/H₀ ≈ 4,300 Mpc. Galaxies farther than this recede from us at apparent velocities exceeding c. This doesn't violate special relativity: it is not motion <em>through</em> space, it is space itself stretching. Relativity forbids objects from travelling faster than light locally — it says nothing about the expansion of spacetime itself.
Only approximately. For H₀ = 70 km/s/Mpc, 1/H₀ = 13.97 billion years — close to but not exactly the true age (13.797 billion). The discrepancy arises because expansion has not been constant: deceleration in the matter-dominated era, then acceleration since ~6 billion years ago under dark-energy influence. In a full ΛCDM model, the true age is computed by integrating H(z) from z=∞ to z=0. But 1/H₀ remains a useful mental order of magnitude.