Table 1a

Assumptions applied in resource mobilisation modelling (see table 1b for source references)

	Unit	Constant	Historic trends	GFF conservative	GFF ambitious	Assumptions and comments	Source
2017 baseline assumptions
2017 resource envelope	$	21 784 338 900				The expenditure in this baseline year on the priority set of RMNCAH-N interventions is equal to the cost of providing current levels of coverage (estimated using List Costing module).	8
Adjustment to % OOP in expenditure	%	50.5				The proportion of OOP was adapted from a source⁴ which estimates the relevant splits for general health expenditure. Given that these particular interventions are more likely to be publicly or donor-funded than general health expenditure, OOP payments were scaled downwards. The c.50% adjustment was derived from evidence for a subset of GFF countries which suggests that about half of OOP payments are used to buy medicines/medical supplies.	7
Split of 2017 expenditure—LMIC (post-OOP adjustment)						The split across different sources of financing (without the OOP adjustment) was based on work from Global Burden of Disease Health Financing Collaborator Network.	4
Domestic	%	53.1
Private prepaid	%	12.4
OOP	%	29.2
DAH	%	5.3
Split of 2017 expenditure—LIC (post-OOP adjustment)							4
Domestic	%	29.3
Private prepaid	%	9.8
OOP	%	20.1
DAH	%	40.9
Domestic resource growth assumptions
GDP growth forecast						The model uses IMF forecasts out to 2022, then follows the approach used in Stenberg et al to forecast GDP trajectories for each country: assuming that real growth rates will converge from their 2017–2022 averages to 2% in 2070. The model is sensitive to GDP growth but, since the GFF does not aim to directly influence GDP, it is treated as exogenous for the purposes of this model. Similarly, the share of government expenditure in GDP is held constant across all scenarios.	1
Convergence year	Year	2070
Convergence growth rate	%	2.0
% GGE/GDP	text	Constant				Constant: The model considers that this share does not change. This may be an underestimation if recent trends continue. For instance, total tax revenue rose from 11% to 14% of GDP in LICs and from 13% to 19% of GDP in LMICs between 1990 and 2012, while revenue growth was generally static in HICs. (Junquera-Varela et al)	1
% GGHE/GGE	Text		Constant	Upwards convergence to median for income-region group; during IC only	Upwards convergence to median for income-region group; tapering off after IC (halving each year)	The share of government budgets spent on health is held constant under the trend scenario (based on analysis of historic trends), and to rise under the conservative and ambitious scenarios. In the ambitious scenario, countries below the median for their income level/regional grouping are assumed to increase that share such that they would catch up by 2030, though progress tails away after their investment case period ends. In the conservative scenario, progress finishes completely at the end of the investment period.	2
↑% priority interventions / GGHE, 2017–2030	pp		1	2	The share of health budgets spent on the priority RMNCAH-N interventions is held constant under the trend scenario and increases by 1 and 2 percentage points (pp) by 2030 under the conservative and ambitious scenarios respectively. The levels of improvement were chosen to give a range of results which were in reasonable proportion to baseline levels (c. 8%).	7
DAH/external resource aligned with the investment case (IC) growth assumptions
	Summary DAH growth assumptions
DAH	Text		Constant	Constant +incremental external IC resources, tapering off after IC	Constant +incremental external IC resources, tapering off after IC	External resources aligned with the investment case (IC) are modelled assuming that a proportion of those resources would be additional relative to the counterfactual (in the sense that they would not otherwise have been allocated to this set of interventions). After the investment case period finishes, the model assumes that additional resources taper away quickly, though with some degree of sustainability.	3
	Investment case period funding
GFF Trust Fund resources available	$	2 600 000 000					6
GFF Trust Fund disbursement timing	Text	Uniform distribution					6
Ratio of GFF Trust Fund resources to other resources during investment case phase							6
IDA/IBRD	Ratio		n/a	4.0	6.0
External	Ratio		n/a	6.0	8.0
Private	Ratio		n/a	1.0	1.5
Adjustment to domestic expenditure for infrastructure costs which may not be included in investment case	%	102					8
	Post-investment case period funding
Growth rate (‘−’ implies declining sustainability after IC ends)							7
IDA/IBRD	%		n/a	−75	−50
External	%		n/a	−75	−50
Private	%		n/a	−75	−50
	Proportion of investment case resources assumed to be incremental (relative to trend)
GFF Trust Fund resources	%	100					8
Other external resources	%		n/a	22	28		8
OOP payment growth assumptions
OOP growth rate assumption	Text	Forecast from literature, by income group				OOP sources of health expenditure are assumed to follow trend growth rates estimated in the wider literature, but are also assumed to reduce in proportion to increases in other sources of health funding. That is, a fraction of every dollar of additional funding mobilised is assumed to replace OOP spending rather than being available to scale-up coverage rates. The coefficient used to characterise the relationship between OOP spending and other funding sources is based on CEPA analysis of estimates from the literature and could benefit from further research.	4
OOP average annualised growth rate, 2015–20 30							4
LMIC	%	4.0
LIC	%	1.5
Elasticity of OOP w.r.t. other resources	%	−8.4					5
Private pre-paid (insurance) growth assumptions
PPP growth rate assumption	Text	Forecast from literature, by income group				Private prepaid sources of health expenditure are assumed to follow trend growth rates estimated in the wider literature, and are otherwise treated as exogenous. Although the GFF may perform some activities encouraging uptake of private health insurance, they are not incorporated in this model.	4
PPP average annualised growth rate, 2015–20 30							4
LMIC	%	4.6
LIC	%	3.8
Efficiency gain assumptions
Efficiency gains (achieved by 2030)						Efficiency gain includes any intervention that reduces the cost of achieving a given coverage rate (efficiency) or which increases the health impact that can be achieved with a given set of resources (effectiveness). For instance, improved alignment around investment cases or better prioritisation of essential interventions could both be represented as efficiency gains. The model expressed these gains as an expansion of the overall resource envelope above and beyond each individual source. Efficiency gains are modelled at different rates of progress over distinct parts of the resource envelope which the GFF could claim to influence. For instance, the GFF Partnership may have large opportunities to improve alignment around investment case resources, but less around non-investment case resources and none around OOP payments.	9
Domestic	%		0.0	2.5	5.0
Private prepaid	%		0.0	2.5	5.0
OOP	%		0.0	0.0	0.0
DAH	%		0.0	2.5	5.0%
Investment case resources (not additive)	%		n/a	6.0	12.0

DAH,development assistance for health; GDP, gross domestic product; GFF, Global Financing Facility; GGE, General Government Expenditure; GGHE, General Government Health Expenditure; HIC, high-income countries; IBRD, International Bank for Reconstruction and Development; IDA, International Development Association; IMF,International Monetary Fund; LMIC, low-income and middle-income countries; OOP, out-of-pocket; RMNCAH-N, Reproductive, Maternal, Newborn, Child and Adolescent Health and Nutrition; CEPA, Cambridge Economic Policy Associates; PPP, Private pre-paid.