How do we handle the special commodity Used and secondhand goods?

[WesIngwersen]

Used and secondhand goods, S00402, are defined by BEA (2009) as "goods that were not produced during the current year, so their value (except for the margin, if any, associated with the sale of used items) is not part of the current-period gross output in the economy. There is no industry that produces used goods; they are commodities only. Nevertheless, used goods are part of supply available for consumption, and they are sold by businesses, government, or households to other sectors. For example, used motor vehicles are not produced by an industry, but they are sold by the business sector to the consumer sector. These transactions result in an increase in PCE and a decrease in PFI (that is, the used vehicles sold by business are netted against new business investment). The negative transaction for business is also viewed as a reduction in the stock of capital available for production."

USEEIO v1 and v2 included Used and secondhand goods,
CEDA does not include Used and secondhand goods,

RESOLVED. Keep Used and secondhand goods for US model. Drop it when integrating into the global model.

[WesIngwersen]

Using the Used goods classification for all types of used goods is not useful for characterizing the emissions associated with used goods - it is too heterogenous. In general however it is important for accounting purposes as its an input into industry output and its important for components of final demand.

[bl-young]

In CEDA, the industry output is recalculated after the Used goods are removed from the Use table

[MoLi7]

Pros for keeping the sector

Keeping special commodities like "Used and secondhand goods" is essential for two key reasons:

1. It maintains the economic balance of the Input-Output Tables (IOTs).
2. It allows for the derivation of correct direct requirement coefficients.

Cons for keeping it

A key challenge in integrating US data into MRIO framework is that many countries' IOTs do not have sectors that correspond to certain U.S. special commodities. Consequently, we must create these sectors for other countries using limited or incomplete information.

My thoughts

It seems sensible to retain the special commodities when constructing the EEIO artifacts for the US, as this is an improvement from CEDA's perspective. However, when integrating the US data into the MRIO model, we could remove these sectors, citing the limited data for other countries and acknowledging the potential for resulting discrepancies in the multi-regional model.