Government Bonds and Balance Sheets: How Primary Dealers Influence Other Banks

In the wake of the European sovereign debt crisis, one question has lingered longer than one might think: how do banks anticipate where government bond yields are going? Markets, of course. Macroeconomic forecasts, too. But there is another source, much quieter and much more interesting: the balance sheets of banks acting as primary dealers—that is, the intermediaries chosen by governments to buy, promote, and distribute public debt. 

This is the subject of new research by Annalisa Prencipe (Accounting Department, Bocconi) and Francesco Grazioli (ESCP Business School, Paris), published in the Journal of Banking and Finance. The study begins with the observation that certain accounting choices made by primary dealersnot only reveal how a bank reports government bonds on its balance sheet but also indirectly reveal what that bank expects regarding future sovereign yields. And when this information is made public in a timely manner, other banks read it, interpret it, and adjust their own behavior accordingly.

A hidden signal in the balance sheets

The context is that of post-2011 European sovereign debt crisis, when authorities began pushing for greater transparency in banks’ balance sheets. In particular, a recommendation by ESMA (European Securities and Markets Authority) encouraged the detailed disclosure of exposures to government bonds. And this is where a seemingly technical detail comes into play: the classification of sovereign bonds. Banks can record them at amortized cost or at fair value. An accounting choice? Yes, but also much more.

As the authors explain, the classification of sovereign debt involves a trade-off between liquidity and profitability. In practice, classifying a security at amortized cost means sacrificing liquidity—it is harder to sell it quickly—while protecting against price fluctuations. And this has profound implications:

“The sovereign debt classification at amortized cost becomes optimal in the event of rising interest rates”

Translation: if a bank expects rates to rise (and thus a decline in security prices), it will tend to use this classification. It is an implicit signal about future expectations.

The domino effect: other banks look and react

This is the heart of the study. Prencipe and Grazioli demonstrate that this signal, when disclosed in financial statements, does not go unnoticed.

“Peer banks divest financial instruments and increase loans when domestic primary dealers disclose more sovereign debt at amortized cost”

In other words, when primary dealers indicate (implicitly) that rates will rise, other banks make two moves:

• they sell financial instruments (to avoid losses in value)
• they increase lending (which benefits from higher rates)

It is a veritable informational domino effect that spreads throughout the banking system.

Data and results: a divided Europe (even in transparency)

The study analyzes over 6,400 bank-year observations from 2012 to 2019 across 19 European countries. A key factor is the difference between countries that adopt detailed disclosure practices and those that do not. Some countries (such as Italy, Spain, and France) make detailed information public, while others (such as Germany and the United Kingdom) remain opaquer.

This difference becomes a natural experiment: where information is public, banks react. Where it is not, they do not.

Even more significantly,

“These effects are more pronounced among peer banks facing greater informational disadvantages”

The less-informed banks—or those with weaker analytical capabilities—are the ones that follow the “signal” from primary dealers the most.

Not just accounting: strategic information

Prencipe and Grazioli’s study overturns a long-held view: accounting choices are not merely representations of the past, but also indicators of the future.

“Sovereign debt classifications are not merely accounting choices but also carry forward-looking information content”

In this sense, financial statements become tools of strategic communication, capable of influencing real-world behaviors such as the provision of credit to the economy.

Transparency, markets, and credit

If accounting disclosure guides investment and lending decisions, then transparency can improve market efficiency but can also amplify imitative behavior and directly influence the supply of credit. In other words, a seemingly technical choice can have macroeconomic effects.

One question remains open: where does the informational advantage of primary dealers come from? The authors admit that the answer is not simple; it could stem from relationships with governments, access to confidential information, or simply greater analytical resources. What is certain is that the banking system observes, interprets, and reacts.