Designing JV Deals for the Future


Download the full version of this article

GettyImages-5988-042161 (2).jpgIT IS CERTAINLY POSSIBLE to negotiate a Joint Venture Agreement without dwelling intensely on the future. Indeed, dealmakers have some very good reasons not to over-prescribe the future. After all, defining the future takes time, draws attention away from getting the deal done, and adds to non-closure risk. Defining the future can also introduce potential liabilities and limit future flexibility, as it may lock the company into commitments that do not make sense down the road. One dealmaker summed it up this way:

"I’m a dealmaker, not a soothsayer. It is dangerous to predict the future – and far wiser to let our executives overseeing the business address events and issues as they arise."

But the low success rates of joint ventures – and high level of post-close misalignment and exit issues – suggest that failing to think deeply about the future may not always be the best way to approach joint venture dealmaking. Consider a few examples where incomplete thinking about the future caused severe and avoidable pain:

  • A U.S. industrial company entered into in a 50:50 JV that consolidated a mature business line with that of a competitor. The exit clause in this deal was structured around a common buy-sell provision, whereby after three years, either party had the right to trigger exit by offering a price at which the non-initiating party could either buy or sell. This seemingly fair exit pricing methodology disadvantaged the U.S. company because it was the “natural seller” of the business. This meant that the counterparty knew that it could offer 10 to 15% below fair market value (FMV) for the business, with little to no fear that U.S. company would choose to be the buyer. That deal term cost the U.S. company $120 to 180 million.
  • A Middle Eastern conglomerate formed a 50:50 JV with a global company to develop and operate a metals processing venture, which included a $10 billion refinery and related infrastructure. The legal agreements were well-written, and included traditional language with regard to joint governance and control. Additionally, the partners entered into mirrored versions of master services and secondment agreements with the venture. Two years into the venture, however, it became obvious that the partners held fundamentally different views of how the JV should be operated vis-à-vis the shareholders. The global partner saw itself as the Lead Operating Partner – that is, the provider of the technology, systems, and practices on which to build and operate the venture – and viewed the venture as an asset within its portfolio that should depend on it for marketing, sales, and other functions. In contrast, the local conglomerate saw the JV as an independent business, which would have its own processes, systems, and brand – and would ultimately build its own marketing capabilities. These fundamental differences in the venture’s operating model caused significant deterioration in trust and management turnover – which then led to delays in investment decisions that cost the shareholders at least $500 million in profits. 
  • A global publishing house entered a joint venture in Asia, seeking to grow in a key emerging market by partnering with an established local player. After three years, the business was not meeting targets, and opportunities were slipping away as competitors gained market share. Yet the JV Agreement prevented either partner from unilaterally replacing the CEO or forcing needed investments. Meanwhile, the exit clause prevented the global publisher from competing in the local market for three years once it exited the JV, so selling was not an option. Ultimately, the negotiated exit agreement was expensive and delayed.
We believe that companies need to expand the set of questions they ask during the deal process – and look out across the full venture lifecycle for more clues about what issues the parties will predictably confront in the JV. Dealmakers should address these issues pre-close, while they have more leverage, and fewer risks. Specifically, front-loading the future means having good answers to a number of questions, including:

  • Does our likely future position disadvantage us relative to the counterparty if we adopt seemingly fair, boilerplate deal terms (e.g., on exit pricing, dividends, future investments, etc.)? 1
  • Have we defined the venture’s strategy, governance, and operating model at sufficient detail to expose any serious future misalignments with the counterparty about how to run the business?
  • Does the proposed deal inappropriately lock us into a partnership from which it will be extremely painful to escape in the future (e.g., should the venture not perform well, or should our strategy change)?
  • Have we adequately anticipated the natural evolution or inflection points of the venture (e.g., the shift from research to commercialization, or from the project/design phase to the build/ operate phase), and driven this into the deal terms?
The purpose of this note is to tackle the first question 2, and to focus on three terms – dividend policies, exit, and future capex. 


Many dealmakers rely on standard terms or model-form agreements to expedite the dealmaking process, and limit the areas of negotiation. But in certain situations, such terms can be highly problematic for one of the partners if certain events occur – events that can be predicted during dealmaking. Therefore, a key element of front-loading the future is looking at these boilerplate terms, and testing whether standard language that seems eminently balanced and fair will actually disadvantage your company down the road. If the answer is yes, then dealmakers need to deviate from the traditional, timeworn path of the boilerplate – either by opting out of standard contractual language, deepening existing language, or developing additional agreements pre-close.

To illustrate this work, consider three deal terms, and how a deeper view of the future drives the company away from the boilerplate:

Dividend policies. The typical JVA requires a supermajority vote of the JV’s Board of Directors (or equivalent) to approve any financial distribution to the shareholders. While this standard term may seem reasonable, it can disadvantage a shareholder in certain situations. For instance, if your company is entering into a 50:50 JV with an eye toward generating near-term financial returns, but the other shareholder is more interested in growth, building market share, or keeping profits within the venture for tax or other reasons, then this standard deal term is bad for you, as it creates a barrier to getting cash out of the business. Simply put: If the partner does not want to make a distribution, venture profits have nowhere to go but to sit in the JV.

Such a standard term can also materially reduce the overall level of capital discipline across a company’s portfolio. Consider a European chemical company, which owned more than ten JVs accounting for some 30% of its total capital spending. A meaningful portion of this capital spending was self-funded by the JVs. “Effectively what this means,” according to the company’s Finance Director, “is that our earnings from these ventures are not ‘re-competing’ for capital with our other businesses and projects, and therefore, our money may not be going to the best investments.”

If your company sees the future unfolding like this, you should consider deviating from the standard JVA voting provisions for dividend policies. As with a number of other standard boilerplate terms, we have developed a checklist of non-standard deal terms to consider, the relevance and feasibility of which will vary by the situation (Exhibit 1). For instance, in the situation described, the company might have proposed that the agreements be structured to provide an automatic payout of any earnings above a certain level (e.g., $10 million, or 20% of the JV’s annual operating costs), unless the Board agrees by unanimous vote to the contrary. Alternatively, you might structure some or all of the shareholders’ initial capital contributions to the JV in the form of a loan, with a pre-agreed repayment schedule linked to JV earnings (and where earnings are automatically used to repay lenders prior to making other capital investments in the venture).

Exhibit 1: Dividend Policies – Beyond the Boilerplate

Exhibit 1 Dividend Policies – Beyond the Boilerplate v3.png

Click to Enlarge

Exit process and pricing. In 50:50 JVs, it is common for the parties to agree to a buy-sell provision as a way to avoid prolonged deadlock, and prevent either party from being stuck in a business that is underperforming, or that no longer fits with its strategy. While there are different types of buy-sell provisions, the basic idea is that either party has the right to initiate a bidding process that will lead to a buyout, but does not know whether it will be the buyer or the seller of the venture. The industrial JV discussed above used a common buy-sell provision often referred to as Russian Roulette.3 Under this structure, either party has the right to trigger exit, with the initiating party offering a valuation at which the non-initiating party is then given the choice to either buy or sell.

On the surface, this pricing methodology seems fair, since the initiating party sets the valuation, but does not know whether it will be the buyer or the seller at that price. However, deeper thinking about the future could lead the company to see that this is a suboptimal structure. Specifically, if your company is the more natural seller of the venture (e.g., because it views the venture as a non-core business, because it is less strong financially than the counterparty, or because the venture has far greater operational integration with the counterparty), then a buy-sell provision disadvantages your company, since the counterparty knows that it can offer below fair market value for the venture.

Download the full version of this article

   By “boilerplate,” we mean contractual arrangements that are fairly standard in Joint Venture Agreements, and are therefore viewed as the default option by dealmakers. Though boilerplate deal terms are often sufficient for satisfying both shareholders’ needs, there are many situations in which creative, non-standard deal terms should be preferred.

2    Other Water Street Partners articles, webinars, and roundtables have addressed or will address the other three questions. For example, on 25 June 2015, Water Street Partners hosted a webinar, “Defining the Governance Model for a JV,” which addressed aspects of the second question. A recording of that webinar is available to our members via the Joint Venture Advisory Group website.

3   Other common forms of buy-sell provisions in JVs include a Texas Shootout and the Dutch Auction (also known as a Mexican Shootout). Under the Texas Shootout, each partner submits a sealed bid containing its perceived value per share of the joint venture, and then the partner with the higher bid must buy out the other partner at that price. Under the Dutch Auction, each partner submits a sealed bid indicating the minimum price per share at which it would be prepared to sell its shares; whichever bid is higher wins, and that bidder then buys the loser’s share at the price indicated in the loser's bid. All three of these common types of buy-sell provisions disadvantage the natural seller.