The historical forecast archive of the World Bank has the following note:
please also note that in the past, commodities forecasts have not appeared in stand-alone publications, but rather as chapters or annexes of other publications, namely Global Economic Prospects (in January of each year) and Global Development Finance (in June of each year)
with links to the respective report archives going back to the late 1990's. So you can add another decade to that dataset by manually scraping the previous reports.
In theory you may be able to use historical futures prices for crude as a proxy for forecasts. Quandl has many energy futures datasets going back up to 50 years. From a brief search, the general conclusion in the literature seems to be that futures prices are slightly better predictors than a random walk most of the time, and much better some of the time. The Federal Reserve, specifically, has addressed this topic a couple of times.
In any case, it depends on what exactly you mean by "oil price expectations rather than real oil prices." If expectations == futures and real == spot, then you are all set.
Finally, my understanding is that Bonny Light is deliverable into NYMEX futures contracts (source):
Specific domestic crudes with 0.42% sulfur by weight or less, not less than 37° API gravity nor more than 42° API gravity. The following domestic crude streams are deliverable: West Texas Intermediate, Low Sweet Mix, New Mexican Sweet, North Texas Sweet, Oklahoma Sweet, South Texas Sweet.
Specific foreign crudes of not less than 34° API nor more than 42° API. The following foreign streams are deliverable: U.K. Brent and Forties, and Norwegian Oseberg Blend, for which the seller shall receive a 30¢-per-barrel discount below the final settlement price; Nigerian Bonny Light and Colombian Cusiana are delivered at 15¢ premiums; and Nigerian Qua Iboe is delivered at a 5¢ premium.
I'm not sure that's always been the case, but it may make your research easier. NYMEX Crude futures have been around since the 1980's.