Identity (6.1) suggests three observations concerning the scope for “opportunistic”
accounting:
a) underestimation (overestimation) of accrued expenditure (revenue) allows
reporting a lower Maastricht deficit (dm) but leaves the change in debt (Δb) unaffected as
it is offset by an increase in cash-accrual differences (CA);
b) similarly, the adoption of loose standards in the identification of
payments/receipts reflecting acquisition/sales of financial assets, reduces the reported dm
but leaves Δb unaffected due to the corresponding increase/decrease in FAa/FAs;
c) sales of financial assets (FAs) and debt restructuring operations (a component of
VE) can be used to reduce the change in debt but have no effect on the Maastricht
deficit.112
Therefore:
a) a large difference between Δb and dm should alert towards the possibility that
dm is underestimated;
b) a small difference between Δb and dm cannot be taken to exclude an
underestimation of dm since sales of assets and debt restructuring can be used to offset
inflated cash-accrual differences and net acquisition of financial assets.
This suggests rewriting (6.1) in order to partition total SFA into two groups. One group
includes items that can be used to affect the Maastricht deficit but leave the change in
debt unaltered (a “deficit-specific” SFA, x), the other includes items that can be used to
reduce the change in debt but leave the Maastricht deficit unaffected (a “debt-specific”
SFA, z):113
Δb ≡ dm + x- z (6.2)
Mapping identity (6.1) into identity (6.2), however, is not a straightforward exercise.
First, there are items in the reconciliation account that do not belong to either x or z (this
is the case of valuation effects arising from fluctuations in the value of foreign currency-
denominated debt and because of bonds issued above/below par).114 Second, some of the
individual items in identity (6.1) may be affected by attempts at reducing dm as well as Δb
(this is the case of asset sales, FAs, whose total can be lowered by an opportunistic
classification of transactions aimed at lowering dm, and can be increased by privatization
programs undertaken to reduce Δb).
112
113
114
Importantly, such operations may leave the government’s net asset position unaffected or even
worsen it. In this respect, one should also control the extent of one-off measures affecting directly dm
(Milesi-Ferretti, 2003). There is of course no implication here that privatizations are by definition
bad policy. However, if they are undertaken with the sole purpose of reducing gross debt -
regardless of the economics underlying the transaction - then the operations can be questioned.
See also Buti et al. (2007).
Note, however, that by issuing bonds above par, a government could reduce the change in debt
associated with a given deficit.
166