When I wrote the full IAS 40 guide, one question kept bothering me. The standard says fair value is its preferred model, and the rebuttable presumption pushes firms toward it. So why does a chunk of European real estate still sit on the cost model, a decade and more after IFRS became mandatory?

I went looking for actual data and found a study that answers exactly that: hand-collected figures from the annual reports of 212 listed European real estate firms, 1,513 firm-year observations across 2014 to 2023, run through a logistic regression. This post is the readable version of what it found.

What the standard expects

IAS 40 hands companies a real choice between two models for investment property, and the standard is not neutral. The fair value model carries a rebuttable presumption that fair value can be measured reliably on a continuing basis. Fair value is the star of the show.

Except reality does not cooperate. If the standard’s preference were the only force at work, you would expect nearly universal adoption of fair value. The study’s first number says otherwise.

The headline numbers

Across the full sample, 84.3% of firm-years use the fair value model. That sounds like a clear verdict, until you split it by country.

  • United Kingdom, Sweden, Finland and Ireland use fair value almost exclusively.
  • Italy and Spain flip it: a majority of firms there still use the cost model.
  • France, Germany and Greece sit somewhere in between.

A decade after IFRS adoption, accounting choices in Europe have not converged on the standard-setter’s favourite model. National accounting traditions survived the transition. Italy and Spain ran on historical cost before IFRS, and a large part of the listed sector there never left it. That divergence is the study’s starting point, and it drives everything below.

How the study was put together

The researchers hand-picked the sample precisely because the countries differ. Nine European countries, chosen so the sample includes jurisdictions that mandated fair value before IFRS (UK, Ireland) and jurisdictions that required historical cost (Italy, Spain, Germany). That contrast is what makes institutional effects measurable.

The dependent variable is binary: 1 if the firm uses the fair value model, 0 if it uses cost. The independent variables test competing explanations:

Variable Measures Theory behind it
Debt ratio (LEV) Total debt to total assets Contractual efficiency: debt covenants and creditor pressure
Firm size (SIZE) Natural log of total assets Political costs: visibility and regulatory scrutiny
Market-to-book (MTBV) Equity value relative to book Information asymmetry between managers and investors
Property materiality (IP_TA) Investment property to total assets Performance measurement and the importance of the asset class
Country institutions Capital market development, political processes Whether national context outlasted IFRS adoption

What the regression found

The full model explains roughly 45% of the variance in the fair value versus cost choice. In accounting-choice research, that is a strong result. Three categories of drivers all come out significant.

Gearing pushes toward cost. The strongest and most clearly directional finding is that firms with higher debt ratios are significantly more likely to use the cost model. The textbook argument carries here: debt holders prefer conservative reporting because an overstated asset base raises the risk of dividend payouts and asset erosion at their expense. The cost model protects creditors, and firms that depend on debt behave accordingly.

Size and market signals matter. Firm size, used as a proxy for the political costs of being visible to regulators and the taxman, plays a role. The study reads it through a real estate lens: unlike many asset classes, investment property markets are often liquid, so their fair values are less speculative than, say, long-dated intangibles. That liquidity weakens the usual objection to fair value.

Information asymmetry and materiality. Market-to-book, standing in for how much managers know that investors do not, and the materiality of investment property relative to total assets both shape the choice. Where the property is a large share of the balance sheet, its valuation genuinely moves the financial statements, so the model choice has real consequences.

Country institutions are the quiet heavyweight. The finding the paper keeps coming back to: a firm’s institutional context, capital market development and political processes, predicts the choice even after all the firm-level variables are in the model. A Greek or Italian company does not price its property like a UK one, and the gap traces back to how the two countries approached valuation before IFRS ever existed.

Why this matters

For standard-setters, the result is uncomfortable. The case for a universal fair value mandate rests on one model serving everyone. The data says the cost model is not a relic or a compliance dodge. It is a rational response to creditor pressure, national tradition and the way capital is raised in certain markets. Ten years post-adoption, firms still negotiate the choice rather than converge on it.

For investors and analysts, the message is to read the accounting policy itself as information. Two identical real estate portfolios can report very different earnings purely because of the chosen model. The choice tells you about the firm’s gearing, its market, and how it sees itself. Look past the face of the financial statements before comparing anything.

For students, there is a practical payoff too. The exam answer to “why would a company use the cost model?” used to be a gesture at conservatism. Now it has numbers behind it: gearing, institutional history, and a country’s pre-IFRS tradition.

What the study does not claim

The honest limits: the sample covers nine European countries and the real estate sector only. That focus buys depth, not breadth. Extending the method to Asian or other IFRS-adopting regions, where institutional histories differ again, is the natural next step. And the model, strong as it is, leaves over half the variance unexplained. Accounting choice is a creature of many small pressures, only some of which show up in a regression.

Key takeaways

  • 84.3% of European real estate firm-years use the fair value model, but that hides a sharp country split: UK, Sweden, Finland and Ireland are near-universal fair value; Italy and Spain still lean cost.
  • Higher gearing is significantly linked to the cost model, consistent with debt holders preferring conservative reporting.
  • Firm size, information asymmetry and the materiality of investment property all shape the choice.
  • Country-level institutions remain a strong driver a decade after IFRS adoption.
  • The full model explains about 45% of the variance, which is high for accounting-choice research.

Why do some European real estate companies still use the cost model under IAS 40?

Higher gearing, political-cost pressure at larger firms, and the institutional history of the home country all pull toward cost. Italy and Spain, which ran on historical cost before IFRS, still show a majority of firms on the cost model.

Which countries prefer the fair value model for investment property?

The United Kingdom, Sweden, Finland and Ireland use fair value almost exclusively. Italy and Spain are the outliers, with a majority of firms still on the cost model.

Does a firm’s debt level push it toward the cost model?

Yes. The regression finds a strong negative and statistically significant association between the debt ratio and the use of fair value, in line with the argument that creditors prefer conservative valuations.

Why does the IAS 40 model choice matter to investors?

Because the choice directly moves reported earnings and balance sheet values. Two portfolios can hold near-identical property and report very different results. Reading which model a firm chose, and why, is part of reading its financial statements.

What did the study examine?

The accounting choice for investment property under IAS 40 across 212 listed European real estate firms, using 1,513 firm-year observations from 2014 to 2023 in nine countries, analysed with a logistic regression.

In closing

The title says it: fair value versus cost under IAS 40 is not a technical formality. It is a negotiated settlement between creditors, markets and national habit, and it shows no sign of settling.

For the full accounting treatment of investment property, from initial recognition to disclosure, the IAS 40 complete guide has the standard section by section. This post is the companion piece: the evidence on which model firms actually pick.