Three Leptons, One Coupling: Universality Tests From Tau Decays

The Standard Model’s electroweak sector has a specific symmetry built into it. The electron, the muon, and the tau enter the Lagrangian through three identical SU(2) doublets, each coupling to the W and Z bosons with the same strength $g$. The only thing distinguishing one lepton flavour from another is the mass, which arises from the different Yukawa couplings to the Higgs boson. The gauge interactions — the coupling to the W in a charged-current process, the coupling to the Z in a neutral-current process — are identical.

This is the principle of lepton universality. It is not a deep symmetry of the underlying gauge structure; it is a feature of the way three nearly-identical doublets were placed in the theory by hand. But because the placement was minimal, any beyond-Standard-Model physics that distinguishes one flavour from another would produce a small but specific deviation from universality at the precision frontier. A few such hints have appeared and faded over the past two decades — most notably the $R(D^{(\ast )})$ tension in B-meson decays, which has weakened with new LHCb measurements — and the search for genuine universality violation continues.

Tau decays are the workhorse of these tests. The tau is the only charged lepton heavy enough to decay, and its 1.777 GeV mass opens up a rich set of final states whose relative rates probe the coupling structure of the three flavours from multiple angles at once. This post is about how tau-decay measurements test lepton universality, what the current bounds are, and what the next generation of experiments (Belle II and LHCb upgrades) will deliver.

What universality predicts

In a charged-current decay of a heavy lepton ${\ell }^{-}\to {\ell }^{\prime -}{\bar{\nu }}_{{\ell }^{\prime }}{\nu }_{\ell }$, the partial width can be calculated exactly at tree level:

$\Gamma (\ell \to {\ell }^{\prime }{\bar{\nu }}_{{\ell }^{\prime }}{\nu }_{\ell })=\frac{g_{\ell }^2{g}_{{\ell }^{\prime }}^2}{(g/\sqrt{2}{)}^4}\cdot \frac{G_F^2{m}_{\ell }^5}{192{\pi }^3}\cdot f(m_{{\ell }^{\prime }}/m_{\ell })$

where $g_{\ell }$ and $g_{{\ell }^{\prime }}$ are the gauge couplings of the two leptons to the W boson and $f(x)=1-8x^2+8x^6-x^8-12x^4\ln x^2$ is a kinematic phase-space factor that depends only on the mass ratio of the two charged leptons. The Fermi constant $G_F=(g/\sqrt{2}{)}^2/(4\sqrt{2}{M}_W^2)$ encodes the W-mediated interaction strength.

If $g_e=g_{\mu }=g_{\tau }$, as the Standard Model requires, then the ratios of partial widths depend only on the kinematic factor and on quantum-electrodynamic corrections. Specifically:

$\frac{\Gamma (\tau \to \mu {\bar{\nu }}_{\mu }{\nu }_{\tau })}{\Gamma (\tau \to e{\bar{\nu }}_e{\nu }_{\tau })}=\frac{f(m_{\mu }/m_{\tau })}{f(m_e/m_{\tau })}\cdot \text{(QED corrections)}$

The ratio is predicted to be approximately 0.9726, with about a 0.1% theoretical uncertainty from the QED corrections. Any measured deviation directly probes universality between the electron and muon couplings to the W.

Similarly, comparing the muon decay rate $\Gamma (\mu \to e{\bar{\nu }}_e{\nu }_{\mu })$ to the corresponding tau decay $\Gamma (\tau \to e{\bar{\nu }}_e{\nu }_{\tau })$ probes universality between the muon and tau couplings.

The three universality ratios

There are three independent ratios that come out of the tau-decay programme:

$g_{\tau }/g_{\mu }$: From the ratio of $\Gamma (\tau \to e{\bar{\nu }}_e{\nu }_{\tau })$ to $\Gamma (\mu \to e{\bar{\nu }}_e{\nu }_{\mu })$, after correcting for the different masses and lifetimes. The current world average is

${\left(\frac{g_{\tau }}{g_{\mu }}\right)}^2=1.0010\pm 0.0014$

consistent with unity at the per-mille level.

$g_{\tau }/g_e$: From the ratio of $\Gamma (\tau \to \mu {\bar{\nu }}_{\mu }{\nu }_{\tau })$ to $\Gamma (\mu \to e{\bar{\nu }}_e{\nu }_{\mu })$. Current world average:

${\left(\frac{g_{\tau }}{g_e}\right)}^2=1.0029\pm 0.0014$

also consistent with unity.

$g_{\mu }/g_e$: From the ratio of $\Gamma (\tau \to \mu {\bar{\nu }}_{\mu }{\nu }_{\tau })$ to $\Gamma (\tau \to e{\bar{\nu }}_e{\nu }_{\tau })$. Current world average:

${\left(\frac{g_{\mu }}{g_e}\right)}^2=1.0019\pm 0.0014$

The three ratios are not independent — they are related by $g_{\tau }/g_e=(g_{\tau }/g_{\mu })\cdot (g_{\mu }/g_e)$ — but the constraints they place on universality-violating new physics are partially independent through the different systematic budgets of each measurement.

What new physics would do

A deviation from universality at the per-mille level would point to specific kinds of new physics. The dominant candidates that produce universality-violating effects include:

Charged Higgs bosons in two-Higgs-doublet models couple proportionally to lepton mass, so the tau coupling is enhanced relative to the muon and the electron. Compared to the Standard Model W exchange, an additional $H^{\pm }$ contribution would shift the ratios $g_{\tau }/g_{\mu }$ and $g_{\tau }/g_e$ upward by amounts depending on $\tan \beta$ and the $H^{\pm }$ mass.

Leptoquarks couple specific flavours of quarks to specific flavours of leptons, with the coupling structure depending on the leptoquark representation. Universality-violating leptoquarks would alter both leptonic tau decays and semileptonic decays differently, producing a characteristic pattern across the universality ratios.

Heavy neutral leptons that mix with the active neutrinos modify the effective PMNS matrix elements and consequently the effective W coupling for each charged lepton. The pattern of deviations follows the heavy-light mixing matrix and overlaps with the PMNS unitarity tests discussed in an earlier post.

Additional W bosons ($W^{\prime }$) with non-universal couplings would directly modify the effective Fermi constants for different lepton flavours. Such bosons are already constrained at the few-TeV scale by direct collider searches and by precision electroweak fits.

Each of these scenarios produces a distinctive pattern across the three universality ratios, so a future improvement in the precision below 0.1% would either reveal a coherent deviation pointing to a specific model or exclude wide regions of parameter space.

The hadronic tau decays

Beyond the leptonic decays, the tau also decays semileptonically: $\tau \to \pi \nu$, $\tau \to K\nu$, $\tau \to \rho \nu$, and many more exclusive and inclusive hadronic final states. These provide additional universality tests through ratios such as

$\frac{\Gamma (\tau \to \pi \nu )}{\Gamma (\pi \to \mu \nu )}\text{and}\frac{\Gamma (\tau \to K\nu )}{\Gamma (K\to \mu \nu )}$

Each ratio probes the universality between the tau coupling at the heavy end and the muon coupling at the light end of the same hadronic vertex. The current world averages place the tau-to-muon ratio consistent with unity at the few-per-mille level, consistent with the leptonic-decay tests.

Inclusive hadronic decays add another precision channel: the ratio $R_{\tau }=\Gamma (\tau \to \nu \text{ hadrons})/\Gamma (\tau \to \nu e{\bar{\nu }}_e)$ is calculable in QCD to about 1% precision and provides a cross-check on the universality fit while also serving as one of the most precise determinations of the strong coupling constant ${\alpha }_s$.

The next generation

Belle II at SuperKEKB has produced enormous tau-pair samples since its restart in 2019, with a target of 50 billion tau-pair events over its full luminosity. The improvement over the previous BaBar and Belle datasets is roughly a factor of 30 in tau-pair statistics, which translates directly into improved precision on the branching ratios.

LHCb, although optimised for B-meson decays, has produced surprisingly competitive tau measurements through its forward-coverage of high-rate tau-pair production. The LHCb upgrade currently in commissioning will increase its tau sample further.

Combined, the next decade should bring the precision on each universality ratio below 0.1%, an order of magnitude improvement over the current world averages. This is the level at which most well-motivated beyond-Standard-Model scenarios become testable, and at which any genuine deviation would become visible.

Why tau universality matters for neutrinos

The lepton-universality tests do not measure neutrino properties directly. But they constrain the structure of the charged-current interaction in a way that propagates to the PMNS sector. If $g_{\tau }\ne g_{\mu }\ne g_e$, then the effective PMNS matrix elements that govern oscillation are not simply rotations of three identical doublets but include additional structure from the universality-violating physics. The non-unitarity tests discussed in the previous PMNS-unitarity post are one specific consequence; the tau-decay universality tests are the complementary direct probe.

In the limit of universality, the Standard Model’s three-flavour neutrino picture is internally consistent. Outside that limit, the relationship between neutrino mixing and the underlying gauge structure becomes more complicated and requires explicit modelling of the new-physics sector that breaks the universality. The fact that universality tests are at the per-mille level and consistent with unity is therefore a small but real constraint on the parameter space for beyond-Standard-Model neutrino physics.

Summary

The Standard Model predicts that the electron, the muon, and the tau couple to the W and Z bosons with identical strengths — the principle of lepton universality. Tau decays provide the most sensitive test of this prediction, because the tau is heavy enough to access both leptonic decay channels and a rich set of semileptonic final states. The current world averages of the three universality ratios — $g_{\tau }/g_{\mu }$, $g_{\tau }/g_e$, and $g_{\mu }/g_e$ — are all consistent with unity at the per-mille level. Belle II’s 50-billion-tau-pair sample, complemented by LHCb upgrades, will push the precision below 0.1% over the next decade. The tests constrain a wide variety of beyond-Standard-Model physics scenarios that distinguish lepton flavours — heavy charged Higgs bosons, leptoquarks, heavy neutral leptons, additional gauge bosons — and complement the direct collider and short-baseline oscillation searches for the same kinds of physics. In the neutrino sector, they bound the structure of the charged-current interaction that governs how neutrinos are produced and detected, and they are part of the broader empirical case for the three-flavour Standard-Model picture being complete at the per-cent level — for now.