MRIOT Input#

Using a IOSystem as input#

The IOSystem given to instantiate a model requires to contain intermediate demand matrix Z, final demand matrix Y and gross production vector x as attributes, and to be balanced.

This should be the case for all MRIOT parsed with the pymrio package. Refer to its documentation for more details.

In particular, if you want to build your own MRIOT based on your own data source, you should read this part of pymrio’s documentation.

Attention

Note that the (region,sector) Multi-indexes for the matrices and vector are reordered by BoARIO to be in lexicographic order.

Productive capital#

The model requires an estimation of the productive capital of each industries in order to model the effect of productive capital destruction. In BoARIO this stock can either be directly given by the user as a parameter to the model, or estimated from the value added computed from the MRIOT using per sectors ratios.

Attention

If no ratios are given, the models defaults to productive capital = 4 * VA [Hallegatte, 2008]. This is a very coarse assumption that you should not rely on.

Attention

The model checks whether the value added computed as source_mriot.x.T - source_mriot.Z.sum(axis=0) (i.e., the gross output minus the intermediate requirements) contains any negative values, and set it to 0. for industries (and raises a warning)

This in turn sets the productive capital for these industries to be null too when it is estimated from VA to capital ratios.

Using a Local Input-Output Table (LIOT)#

So far the model has not been tested with LIOT. While it doesn’t pose any problem conceptually, the shape of the data might not fit the model.

We suggest the user defines a two-regions MRIOT with one region being the local one, and the other being the Rest of the World made of the exports and imports flows.

Feedback on successively running the model in such setup would be greatly appreciated!