14.

Goodwill


Million US dollar	31 December 2018			31 December 2017
Acquisition cost
Balance at end of previous year		140 980			135 897
Effect of movements in foreign exchange		(7 541	)		4 684
Disposals through the sale of subsidiaries		(652	)		—
Acquisitions through business combinations		107			398
Hyperinflation monetary adjustments		435			—
Reclassified as held for sale		(13	)		—

Balance at end of the period		133 316			140 980
Impairment losses
Balance at end of previous year		(40	)		(34	)
Impairment losses		—			(6	)
Disposals through the sale of subsidiaries		35			—

Balance at end of the period		(5	)		(40	)
Carrying amount
at 31 December 2017		140 940			140 940
at 31 December 2018		133 311			—

On 30 March 2018, AB InBev completed the 50:50 merger of AB InBev’s and Anadolu Efes’ existing Russia and Ukraine businesses. Following this merger, the company derecognized its Russian and Ukrainian net assets including goodwill (see also Note 6 Acquisitions and disposals of subsidiaries).

The carrying amount of goodwill was allocated to the different cash-generating units as follows:


Million US dollar Cash-generating unit	2018		2017
United States		33 288		33 277
Colombia		18 802		20 425
South Africa		15 896		18 551
Peru		14 513		15 074
Mexico		12 614		12 580
Rest of Africa		7 716		8 326
Australia		6 348		6 922
Brazil		4 715		5 523
South Korea		3 949		4 119
Ecuador		3 925		3 925
China		2 758		2 914
Honduras & El Salvador		2 284		2 335
Canada		1 891		2 056
Other countries		4 613		4 913

Total carrying amount of goodwill		133 311		140 940

AB InBev completed its annual impairment test for goodwill and concluded that, based on the assumptions described below, no impairment charge was warranted.

The company cannot predict whether an event that triggers impairment will occur, when it will occur or how it will affect the value of the asset reported. AB InBev believes that all of its estimates are reasonable: they are consistent with the company’s internal reporting and reflect management’s best estimates. However, inherent uncertainties exist that management may not be able to control. During its valuation, the company ran sensitivity analysis for key assumptions including the weighted average cost of capital and the terminal growth rate, in particular for the valuations of the US, Colombia, South Africa, Peru and Mexico, countries that show the highest goodwill. While a change in the estimates used could have a material impact on the calculation of the fair values and trigger an impairment charge, the company, based on the sensitivity analysis performed is not aware of any reasonably possible change in a key assumption used that would cause a cash-generating unit’s carrying amount to exceed its recoverable amount.

Goodwill impairment testing relies on a number of critical judgments, estimates and assumptions. Goodwill, which accounted for approximately 57% of AB InBev’s total assets as at 31 December 2018, is tested for impairment at the cash-generating unit level (that is one level below the operating segments). The cash-generating unit level is the lowest level at which goodwill is monitored for internal management purposes. Except in cases where the initial allocation of goodwill has not been concluded by the end of the initial reporting period following the business combination, goodwill is allocated as from the acquisition date to each of AB InBev’s cash-generating units that are expected to benefit from the synergies of the combination whenever a business combination occurs.

AB InBev’s impairment testing methodology is in accordance with IAS 36, in which fair-value-less-cost-to-sell and value in use approaches are taken into consideration. This consists in applying a discounted free cash flow approach based on acquisition valuation models for its major cash-generating units and the cash-generating units showing a high invested capital to EBITDA multiple, and valuation multiples for its other cash-generating units.

The key judgments, estimates and assumptions used in the discounted free cash flow calculations are generally as follows:

•

In the first three years of the model, free cash flows are based on AB InBev’s strategic plan as approved by key management. AB InBev’s strategic plan is prepared per cash-generating unit and is based on external sources in respect of macro-economic assumptions, industry, inflation and foreign exchange rates, past experience and identified initiatives in terms of market share, revenue, variable and fixed cost, capital expenditure and working capital assumptions;

•

For the subsequent seven years of the model, data from the strategic plan is extrapolated generally using simplified assumptions such as macro-economic and industry assumptions, variable cost per hectoliter and fixed cost linked to inflation, as obtained from external sources;

•

Cash flows after the first ten-year period are extrapolated generally using expected annual long-term GDP growth rates, based on external sources, in order to calculate the terminal value, considering sensitivities on this metric;

•

Projections are discounted at the unit’s weighted average cost of capital (WACC), considering sensitivities on this metric;

•

Cost to sell is assumed to reach 2% of the entity value based on historical precedents.

For the main cash generating units, the terminal growth rate applied generally ranged between 1% and 4%.

The WACC applied in US dollar nominal terms were as follows:


Cash-generating unit	2018			2017
United States		7	%		6	%
Colombia		7	%		7	%
South Africa		8	%		8	%
Peru		7	%		7	%
Mexico		8	%		9	%
Rest of Africa		11	%		10	%
Australia		7	%		6	%
South Korea		7	%		6	%
Ecuador		11	%		11	%

In the sensitivity analysis performed by management, an adverse change of 1% in WACC would not cause a cash-generating unit’s carrying amount to exceed its recoverable amount.

The above calculations are corroborated by valuation multiples, quoted share prices for publicly-traded subsidiaries or other available fair value indicators (i.e. recent market transactions from peers).

Although AB InBev believes that its judgments, assumptions and estimates are appropriate, actual results may differ from these estimates under different assumptions or market or macro-economic conditions.