Hello all, I am reading through Friedland's textbook on loss reserving. There is one question. When using the development triangles to summarize the data, I never see the textbook to adjust the inflation.
For example, if an accident year 2000 claim X got paid 1000 in 2000, and further paid 1150 in 2001, the textbook always (up to where I am reading now, which is Cape-cod technique) put 1000 + 1150 = 2150 into the entry corresponding to development year 1 for accident year 2000.
This really looks weird for me. If for a specific calender year there has been an exceedingly level of inflation applicable to a LOB (not necessarily applicable to the overall economy), say it is inflation of 15% for just 2001, then this 1150 should be 1000 in 2000 dollars. Furthermore, if this 15% decreases back to its normal level one year later, one can see that all the 2001 calendar payments are inflated too much. This will perhaps cause some problems to LDF.
What I think is, which maybe makes more sense in such circumstances, to deflated everything when constructing the development triangle. For example, you can express all the incremental payment (year-to-year payment) first in terms of 2000 dollars, and then sum them up to get a cumulative payment, the resulting LDF is free of the influence of any inflation.
Why are they not doing this kind of deflation of dollars in the textbook? Thanks for comments and advises.
For example, if an accident year 2000 claim X got paid 1000 in 2000, and further paid 1150 in 2001, the textbook always (up to where I am reading now, which is Cape-cod technique) put 1000 + 1150 = 2150 into the entry corresponding to development year 1 for accident year 2000.
This really looks weird for me. If for a specific calender year there has been an exceedingly level of inflation applicable to a LOB (not necessarily applicable to the overall economy), say it is inflation of 15% for just 2001, then this 1150 should be 1000 in 2000 dollars. Furthermore, if this 15% decreases back to its normal level one year later, one can see that all the 2001 calendar payments are inflated too much. This will perhaps cause some problems to LDF.
What I think is, which maybe makes more sense in such circumstances, to deflated everything when constructing the development triangle. For example, you can express all the incremental payment (year-to-year payment) first in terms of 2000 dollars, and then sum them up to get a cumulative payment, the resulting LDF is free of the influence of any inflation.
Why are they not doing this kind of deflation of dollars in the textbook? Thanks for comments and advises.
why we don't adjust for inflation in development triangles?
0 commentaires:
Enregistrer un commentaire