The Agency problem
Small companies are typically managed and owned by the same person. This structure is ideal in the sense of shareholder value maximisation. There is a direct incentive for the manager to maximise wealth for the shareholder as they are one and the same person.
In large companies this structure is usually unobtainable and/or not feasible. Large amounts of capital are required to fund large companies and a single owner would find it difficult to raise such funds. Therefore, the funds must be raised from multiple outside sources meaning a larger number of owners.
As there are many owners in large companies they cannot all participate in the management of the firm. The owners elect a board of directors who in turn oversee the management of the company. The management (or controllers) of the firm and the owners of the firm are then two separate entities.
The upside of this structure is that the owners have the flexibility to buy or sell their stake in the company as and when they see fit. This change in ownership does not affect the day-to-day running of the company. The downside is the incentive for management to maximise shareholder value is not direct and hence diversions can arise. Managers may place their own interests above the shareholders and not maximise shareholder value. This is the agency problem.
Jensen & Meckling (1976) argue that if a manager is faced with a project that will benefit themselves they will undertake the project (assuming the manager holds no equity in the firm), whether the project does or does not benefit the shareholder. This results in a misallocation of current resources and future capital. Current shareholders will not receive efficient value and therefore they (and future potential shareholders) will not allocate capital to the firm.
Jensen, M., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3, 305-360.