Table of Contents Table of Contents
Previous Page  11 /304 Next Page
Information
Show Menu
Previous Page 11 /304 Next Page
Page Background

11

臺大管理論叢

28

卷第

1

at a higher price than the fair value from their investees, or borrow money at higher

interest rates from their controlled entities.

In addition, the probability of detection can reduce with the length of corporate

pyramid. Regulators often have concerns about audit deficiencies that are attributed to the

business group setting (Doty, 2011). While many studies find tax-motivated income

reallocation among the affiliated firms (Harris, 1993), and firms that have more entities,

such as subsidiaries have greater opportunities for tax avoidance (Mills, Erickson, and

Maydew, 1998), the use of corporate pyramid makes it increasingly difficult for outsiders

to disentangle the rent extraction and tax avoidance. Bushman, Chen, Engel, and Smith

(2004) suggest that organizational complexity can limit corporate transparency and

increase information asymmetry. Manconi and Massa (2009) and Chan and Hsu (2013)

argue that the number of layers within a corporate pyramid makes the firm more opaque

and more difficult for the capital market participants to access its information. For

example, to transfer financial information from the bottom of investment layers to the

parent firm, the information of Shanghai Pacific Department Stores at layer 5 (See Figure

1) is consolidated with the parent’s accounting information through 5 times consolidation

process. Combining long length of layers of subsidiaries within a pyramidal firm creates

information aggregation problems that can result in substantial information asymmetries

within a firm, or between insiders and outsiders (including tax authorities). As the number

of investment layers increases, the process of transferring information from the lowest-tier

subsidiaries to the parent company becomes more complex, thereby rendering it more

difficult for external investors to assess and monitor the risk in the investment projects of a

firm.

With the above reasons, we expect that the number of layers is positively associated

with corporate tax avoidance. We form our hypothesis, stated in its alternative form, as

follows:

H1: Ceteris paribus, corporate tax avoidance is positively associated with the

number of layers in corporate pyramids.

We next examine whether the use of corporate pyramid makes it even easier to avoid

tax through offshore havens. Prior studies argue that multinational firms reallocate taxable

income from high-tax jurisdiction to low-tax-jurisdiction, or tax havens and the

reallocation occurs through structuring transactions between affiliates, such as royalty

payments, dividend repatriations and intra-firm debt (Dyreng and Lindsey, 2009;