Evidence & data

Charts and graphics

The book's principal figures, drawn for the web with their sources and limitations attached.

A small number of the book’s figures are redrawn here for the web. Each one carries its source, units, coverage, period, the transformations applied and the things it cannot show — the same standard the book asks the sector to meet, applied here first.

Charts appear in place, on the tour page for the chapter that uses them, rather than in a separate gallery.

Published figures

The convenient number — the headline tuition fee against its real value in 2012–13 prices, using the government’s own published estimates rather than a deflator applied here. How Is It A Market If The Price Can’t Change?

The two deficits, and what covers them — teaching and research cost recovery from the Annual TRAC return, and the income asked to bridge the gap. The Cross-Subsidy Sleight of Hand · When Costs Rise but Income Cannot · Fuelling The Prestige Furnace

The demographic bulge — the size of the domestic eighteen-year-old cohort over time, and what it does and does not guarantee. How a University Runs Out of Road

Move the dial — international fee dependency, and how quickly a change in recruitment moves an institution’s position. How a University Runs Out of Road · The Cross-Subsidy Sleight of Hand

On modelled figures

Several further datasets behind the book involve projection rather than observation. Those are not published as charts here: a modelled series shown next to observed data invites a confidence it has not earned. Where the book uses a projection, the text says so and the reasoning is given there.

The policy simulator

The one exception is deliberate and kept apart. Can you fix the universities? is an interactive model of English higher education’s finances to 2036, in which you change the rules and watch 279 synthetic providers respond. Every result is labelled as illustrative, no real institution is modelled or named, and every assumption, its evidence grade and its sources are published, together with how well the model reproduces published figures and what it leaves out.