Page 161 - Contributed Paper Session (CPS) - Volume 1

CPS1239 Valerie M.B. et al.
and not all statistical offices adjust appropriately; (ii) “free” goods, such as
Facebook, Wikipedia, pictures from a phone, etc., are not included in national
accounts (because their price is zero), thereby underestimating the value they
contribute to GDP. If these platforms are used for e-commerce, for example
(which is very common in developing Asia), their contribution to efficient
distribution is not properly accounted for; (iii) goods or services produced but
not remunerated (unpaid household work, family help) are also not included
because they are free; (iv) when corporations splinter production offshore, the
valuation of each of the stages of production sometimes relies on inaccurate
pricing by multinational companies, who declare their ownership of each stage
of production in the locality that minimizes their tax liability (transfer pricing).
Even if all production stages could be accurately valued, it would require all
countries providing full, accurate reporting and sharing their data on
companies with other national accounts statistics offices, which is beyond the
capacity of most countries’ institutions (Moulton and van de Ven 2018); and
(v) the spillover effects from agglomeration economies of a talented team
working together to produce new knowledge is crucial to productivity and
generally not accounted for. The human capital of a university scientist in the
team, for example, is classified as an “education” service. Such a service is
valued at cost—sometimes subsidized if provided by the public sector—
because there is no tangible output.

2. Methodology and evidence using input-output data
The ADB Multi-Regional Input-Output Table (MRIOT) allow us to measure
sector-level components of servicification using some refinements on the
well-known direct and Leontief coefficients. Using the technical coefficient
matrix, we can quantify the number of services that are directly used as inputs
in manufacturing sectors for arm’s-length transactions. By subtracting this
matrix from the Leontief matrix, we also obtain an estimate of services that are
indirectly used by a particular sector (see ADB (2018) for a detailed description
of the decomposition). The Leontief coefficients themselves give us the total
number of services used in manufacturing. In other words, they represent the
sum of what we denote as direct and indirect components. To illustrate these
concepts, consider the case of an automobile manufacturer. To produce one
vehicle, it uses equipment leased by another company. The rent paid for the
equipment is an example of a direct service used as an input by the automobile
manufacturer. However, this does not account for all the equipment rentals
that are paid in the production of one vehicle. For instance, the automobile
manufacturer may require basic metals as part of its raw materials. Assuming
these metals are also produced using leased equipment, then the rent serves
as an indirect input to the manufacture of a vehicle. Figure 8 shows that the
direct contribution of services to manufacturing’s value added between 2000

150 | I S I W S C 2 0 1 9