I am going over Multinational Operations and things aren’t adding up on their huge example they run through on the CFAI curriculum text. If you look at Vol. 2, p. 173, on the bottom, they compute the translation adjustment under the temporal method. They multiply the "Loss on Dec. 31, 2000 liability as LC 171 [(1/1.06) - (1/0.95)]= 19, which is saying that they are taking the net liability and multiplying it by the change in year in year end exchange rates. I understand that but the numbers, if you look back at exhibit 2, indicate that they are taking the year end rate for 2000 and subtracting from that the year end rate for 2001. Shouldn’t they be taking the difference between 1999 and 2000. Seems like a major error in the book and one that is continued through the whole example. Or maybe I am just not using my algebra correctly. Guidance on this issue would be much appreciated.

also, if you look at the Loss on 2001 increase in liability, they take (1/1.02, which is the average rate) and subtract from that (1/0.95, which is the year end rate). Shouldn’t it be the other way around?

Regarding which year end rates to use: When they mutliply LC171 by the change in the reciprocal of the exchange rates, they correctly use the 2000 and 2001 rates. This is because they are calculating the loss for 2001. They would use the 1999 and 2000 rates only if they were calculating the loss for 2000. Regarding which rate to subtract from which: If you were to use their calculation of [(1/1.06 ) - (1/0.95)] you get a negative number. If you multiply that negative number times LC171, you get a -19. The Exhibit shows a positive 19. So that appears to be an error. Since we’re computing the change in Liabilities, a POSITIVE number would mean a translation LOSS because we have more liabilities. I suppose that they could argue that they are computing the translation loss directly and not the change in liabilities. If that is the case, then there formula is correct…misleading and confusing, but correct nonetheless. So the short answer is that they use this calculation because they are calculating liabilities. If we were performing these calculations on assets, then we would use your calculation of [(1/0.95) - (1/1.06)]. This is because assets are good, so an increase in assets would mean a translation gain. In fact, that’s what they did in Example 2 which begins on page 188. Example 2 involves calculation of assets. So I guess the reasoning behind all of this is that you use two different calculations so that a negative number always means a loss and positive number always means a gain. Personally, I think we’re smart enough to apply the context to determine the correct result, but the authors apparently disagree. I hope this helps. LobsterBoy