Asset sales are a means of financing
One might think that asset sales are motivated not by a need for financing, but rather by a desire to reorganise the firm. This view would suggest that assets should be sold to any buyer that can operate the asset more efficiently than the seller. However, survey evidence indicates that managers also view asset sales as a method of raising capital, and as an alternative to issuing securities like debt or equity (Hite et al. 1987, Borisova et al. 2013). Anecdotally, BP has targeted $45 billion in asset sales to meet obligations arising from the Deepwater Horizon disaster, and banks such as BNP and Société Générale have sold billions of dollars of assets to recapitalise their balance sheets during the financial crisis. Moreover, empirical studies show that the funds raised from asset sales are used for investment and R&D (Hovakimian and Titman 2006, Borisova and Brown 2013). All of these examples suggest that asset sales are often undertaken to raise capital, and not merely to transfer assets to the optimal owner.
Explaining the use of asset sales for financing
Why would a firm raise capital by selling assets instead of issuing securities like debt or equity? Existing research offers some clues. Myers and Majluf (1984) demonstrate that managers raise capital using the claim that is subject to the least amount of information asymmetry about its value, due to the Akerlof (1970) ‘lemons’ problem. Outside investors are less informed than the manager about the quality of a firm’s assets. Since the value of debt is less sensitive than equity to the quality of a firm’s assets, they are more willing to buy debt than equity. One might similarly expect that asset sales are sold if and only if they exhibit the least information asymmetry of the available financing methods. Firms should thus sell assets only if they have an asset that outsiders find easy to value.
In Edmans and Mann (2013), we demonstrate that this intuition is incomplete. In our model, firms possess a core and a non-core asset, and face a financing need which they can meet either by selling the non-core asset or by issuing an equity claim on the firm’s entire balance sheet. The values of the firm’s assets are private information to the manager, and so we do have the effect anticipated by Myers and Majluf – firms are more likely to sell non-core assets when those assets exhibit relatively less information asymmetry than core assets. However, three additional factors also affect the financing decision and can be more important:
- The certainty effect: An equity claim includes ownership of not only the firm's assets, but also the new capital that the firm raises. For example, if the firm raises $1bn of new equity, a new shareholder has a claim not only to the existing assets (whose value is uncertain), but also the $1bn of funds injected (whose value is certain). This certainty effect mitigates the uncertainty surrounding the value of assets in place, particularly when the financing need is high (e.g. if the firm raises $2bn rather than $1bn). Thus, the choice of financing method depends on the amount being raised, and not just the information asymmetry of the firm’s assets. In turn, the size of the financing need has real effects on firm organisation, as relatively small financing needs lead to a greater frequency of asset sales. Interestingly, the certainty effect continues to hold even if the funds raised are used for a risky investment rather than held as cash.
- The camouflage effect: When a firm chooses to sell a non-core asset, this can mean one of two things. It may be that the asset is inherently low-quality (a ‘lemon’ as in Akerlof 1970). Alternatively, it could be that the asset is inherently high-quality, but does not fit well with the seller’s business strategy (is dissynergistic), and thus is more efficiently operated by another firm. This camouflage means that outsiders do not automatically conclude that an asset being sold is low-quality, and are thus rationally willing to pay a reasona