Calculate investments ex ante

Why capital allocation is not a decision-making process, but a calculation problem

In most organizations, investment decisions are understood as a management or governance process: Projects are evaluated, prioritized, discussed and finally decided.

This view is structurally incomplete.

In reality, every investment decision is a combinatorial optimization problem. The central question is not: Which project is good? It is: Which combination of projects maximizes the overall benefit under real restrictions?

With each additional investment option, the decision space grows exponentially (2^N). With just 30 projects, there are over 1 billion possible portfolios. As a consequence, classic decision-making processes inevitably reduce this space and make decisions based on incomplete considerations.

Ex-ante calculation therefore means: no longer selecting, but calculating.

Executive Summary

In many companies, public institutions and capital-intensive organizations, investment decisions are still treated as if they were primarily a question of experience, management quality and good coordination. Projects are evaluated, business cases drawn up, risks discussed, priorities set and budgets allocated. This process is established, comprehensible and organizationally compatible. It creates structure, transparency and the ability to act.

However, this is precisely where its limits lie.

In reality, the actual investment decision does not concern individual projects, but rather the selection of an optimal combination of projects under limited budgets, scarce resources, strategic targets, upper risk limits and operational dependencies. This means that the investment decision is not a linear prioritization problem, but a mathematical decision problem in combinatorial space.

Simply prioritizing investments does not solve this problem. Those who merely evaluate projects describe quality, but do not yet calculate an optimal allocation. Those who only discuss do not replace analysis with knowledge, but often with reduction. The result is decisions that seem plausible but remain structurally suboptimal.

Ex ante calculation means mathematically determining investment decisions before they are implemented. Not to explain ex post why a portfolio has or has not worked, but to calculate ex ante which project combination under the given restrictions will generate the highest degree of target achievement, the highest return on investment or the greatest strategic impact.

For decision-makers, this means a paradigm shift: away from the question of which individual project is convincing and towards the question of which overall combination represents the global optimum under real conditions.

The basic problem of the investment decision

Most investment processes have historically evolved from governance and planning logics. This is understandable. Organizations need procedures for submitting, structuring, comparing and transferring projects to decision-making rounds. This gives rise to portfolios, committees, approval stages, budget discussions and prioritization logics.

These processes fulfill important functions. They create order. They increase communicative clarity. They reduce organizational friction. They enable accountability. However, they are not identical with an optimal investment decision.

The central error in thinking is that organizations often act as if the selection consists of individual good or bad projects. In fact, it consists of possible combinations. A project is not optimal because it is highly attractive in isolation. It is only part of an optimal decision if its inclusion in combination with other projects generates the highest overall benefit under real restrictions.

This is precisely where the mathematical dimension of capital allocation begins.

If an organization has 10 investment options, there are not 10 possible decisions, but 2^10, i.e. 1,024 possible portfolios. With 20 options, there are already 1,048,576 combinations. With 30 projects, the space exceeds 1 billion possibilities. With 50 projects, it is 1,125,899,906,842,624 combinations. This order of magnitude can no longer be fully mastered intuitively, discursively or heuristically.

This makes it clear that the real difficulty of investment decisions lies not in the evaluation of individual projects, but in the structure of the entire decision space.

Why classic methods reach their structural limits

Traditional decision-making processes almost always work with reduction. This reduction makes sense from an organizational point of view, but is mathematically problematic. Projects are pre-selected, divided into categories, separated by department, reduced to scorings or decided sequentially. Options are ruled out in committee meetings, budgets are reserved in advance and initiatives are released step by step.

Each of these steps reduces complexity. This is often exactly what is intended. At the same time, however, it artificially restricts the scope for decision-making.

This has a decisive consequence: the best combination of projects may no longer be considered at all. Not because it is bad, but because it was not visible in the structural pre-selection. The organization then no longer decides between all relevant options, but only between a reduced number of alternatives.

In practice, this often seems sensible. It saves time, reduces discussion effort and creates apparent clarity. However, from an economic point of view, this can be the core of misallocation. This is because suboptimal decisions are not only the result of poor projects, but above all of a mathematically incomplete selection architecture.

This is particularly true when there are interdependencies between projects. Some projects only develop their full value in combination with others. Some projects crowd each other out. Some create synergies. Some increase the impact of another project indirectly via resources, capacities or strategic connectivity. Such interrelationships can only be depicted to a limited extent with simple prioritization lists.

From the decision to the calculation

The classic logic is: we make a decision and check later whether it was the right one. This logic is ex-post oriented. It evaluates results retrospectively. It is capable of learning, but not necessarily of optimizing.

Ex ante logic works differently. It does not ask after implementation whether the portfolio decision was sensible, but calculates the best achievable combination under the available data, objectives and restrictions before the decision is made.

This is not a semantic difference, but a structural one. Justification dominates ex post. Ex ante dominates calculation. Results are interpreted ex post. Ex ante, options are systematically compared. Ex post, management is often an observer of the result. Ex ante, management becomes the designer of the decision-making model.

This also shifts the management question. It is no longer primarily: Which projects do we like strategically? But rather: Which targets, constraints and priority weightings do we define so that a mathematically robust optimal decision can be calculated?

The importance of the complete investment list

One point that is often underestimated is the completeness of the underlying investment options. Ex-ante calculation can only determine the global optimum if the relevant options are available in full before the decision is made.

This sounds trivial, but it is not.

In many organizations, investment decisions are made step by step. Projects are introduced one after the other, informally supplemented, politically delayed or only become visible at a late stage. This means that the decision-making space is constantly changing. The organization then does not optimize across all relevant options, but across a random or historically grown section.

Mathematically, this means that the calculated result can at best be the optimum within an incomplete search space. The global optimum remains invisible if parts of the decision space have not been included.

For professional capital allocation, the complete investment list is therefore not an administrative side issue, but a structural prerequisite. Only when all relevant investment options, alternatives, expansions, scaling or project variants are available can an optimization answer the question of which combination is actually superior.

Why fixed assets are particularly critical

The relevance of ex-ante calculations is particularly clear when it comes to investments in fixed assets. These include infrastructure projects, production facilities, machinery, real estate developments, energy infrastructure and long-term modernization programmes.

These investments are characterized by high capital commitment, long terms and limited reversibility. Those who make the wrong allocations here are not simply correcting a quarterly result, but perpetuating structural inefficiencies over years or decades. Once capital has been tied up, it can only be reallocated with a high degree of friction. This is precisely why the quality of the decision before implementation is so important.

In fixed assets, it is not enough to identify individual projects with positive business cases. Several investments that make sense on their own can still be worse in combination than an alternative portfolio with a different budget structure, different time horizons or a different risk profile. Political or area-related equal distributions also often lead to portfolios that are compatible within the organization but not economically optimal.

The greater the irreversibility of the capital commitment, the greater the leverage of an ex-ante calculated allocation.

What the global optimum actually means

The term global optimum is often used, but rarely clearly defined. It refers to the combination of investments that generates the highest degree of target achievement among all permissible combinations. Permissible means: within defined restrictions such as budget, capacity, risk, personnel availability, minimum strategic requirements, regulatory limits or logical dependencies.

What is important here is that a global optimum is not an opinion. It is not a political majority decision. Nor is it a management preference in the traditional sense. It is a property of the modeled decision space.

This does not mean that the organization no longer plays a normative role. On the contrary: the organization defines the target values, weightings and restrictions. In other words, it determines what is to be optimized. Mathematics then answers which combination is optimal under these specifications.

This is precisely the difference between governance and calculation. Governance defines the framework. Calculation determines the best solution within this framework.

Local rationality is not global rationality

Many bad decisions in portfolios are not the result of irrationality, but of local rationality. Projects are rationally evaluated individually. Every approved project can be justified. Each area receives a comprehensible budget. Each priority can be explained in terms of arguments. And yet the overall result remains suboptimal.

Why? Because good local decisions do not automatically lead to an optimal global combination.

A project with a high isolated ROI can be part of a worse overall solution than several projects with a slightly lower individual value that together generate a higher impact. A strategically attractive project can block resources that would have had a disproportionately high impact elsewhere. A seemingly small intervention in the budget allocation can lead to a completely different and significantly better portfolio structure.

These correlations cannot be reliably identified by intuition. The combinatorial space is too large for this and the interaction between projects is too complex.

Why scoring, prioritization and simulation are not enough

Scoring models, utility value analyses and prioritization matrices are useful tools. They help to structure criteria, make quality visible and systematize discussions. They are often a useful input. But they are no substitute for optimization.

The reason is simple: a score evaluates a project. An investment decision, however, selects a portfolio. There is a methodological leap between the two.

Even simulations do not fully solve this problem. Monte Carlo analyses, sensitivities or scenario calculations can depict uncertainty and examine robustness. They show what could happen under different assumptions. However, they do not automatically answer the question of which investment combination is the best under the given conditions.

Simulation is descriptive. Optimization is crucial.

Those who only simulate understand possible futures better. Those who optimize calculate the best decision within these future assumptions. Both have their place. But they should not be confused.

The economic core: capital allocation as a value driver

The economic leverage of ex-ante calculated investment decisions is considerable. In many organizations, financing costs, process optimization, pricing strategies or productivity are intensively discussed. At the same time, the biggest lever often remains underexposed: the quality of upstream capital allocation.

Even small improvements in the portfolio composition can have significant effects on ROI, EBIT, risk structure, liquidity development and strategic target achievement. The reason for this does not lie in magical individual projects, but in the fact that in reality capital is almost always faced with several competing uses.

If capital is allocated suboptimally, opportunity costs arise. These are often invisible because the alternative better combination was never realized. This is precisely why misallocation is often underestimated. It does not appear as an error in the classical sense, but as a seemingly normal result of a plausible decision-making process.

Ex-ante calculation makes this hidden difference visible. It not only shows what was chosen, but also what would have been better under the same restrictions.

Table: Classic investment logic vs. ex-ante calculation

Dimension	Classic investment decision	Ex-ante calculation
Decision object	Individual projects	Project combinations
Methodology	Evaluation, prioritization, discussion	Combinatorial optimization
Dealing with complexity	Reduction and pre-selection	Systematic calculation of the decision space
Decision logic	Heuristic and discursive	Mathematical and model-based
Result	Plausible selection	Optimal permissible combination
Role of management	Selection and justification	Definition of objectives and restrictions
Transparency about opportunity costs	Limited	Explicitly derivable
Suitability for high complexity	Limited	High
Typical optimum	Local optimum	Global optimum

What is changing for executives

For board members, managing directors, CFOs, investment committees and public decision-makers, ex-ante calculation not only changes the methodology, but also the self-image of leadership. The core performance of management no longer primarily consists of subjectively selecting the best projects from a reduced list. It consists of defining a clean decision-making model.

This includes four questions in particular:

Firstly, which target figure should be maximized? Return on investment, impact, strategic contribution, robustness or a weighted multi-objective system?

Secondly, which restrictions are real and binding? Budget, personnel, time, risk, regulatory limits, minimum quotas or dependencies?

Thirdly: Which investment options are complete and ready for decision?

Fourthly: Which model logic ensures transparency, traceability and governance capability?

Management thus shifts away from intuitive case-by-case decisions and towards the design of a robust decision-making framework. This increases objectivity without removing the responsibility of management. On the contrary: it professionalizes it.

Calculating ex ante does not mean ignoring complexity

Some objections to mathematical optimization are based on the misunderstanding that ex ante calculation is intended to simplify reality or abolish management judgement. The opposite is true. Good optimization does not reduce reality to a naive key figure, but translates real restrictions, conflicting goals and strategic preferences into an explicit model.

This does not suppress complexity, but formalizes it. Uncertainty does not disappear, but can be modeled. Conflicting objectives are not treated rhetorically, but their effect on the portfolio decision is made visible. This is precisely a qualitative advance over decision-making logics in which key assumptions remain implicit, political or situational.

Conclusion

In terms of their structure, investment decisions are neither a pure management issue nor a mere prioritization issue. They are a mathematical allocation problem. Those who ignore this reduce complexity organizationally instead of mastering it analytically.

The central economic question is therefore not whether a project is good. It is which permissible combination of projects generates the highest overall benefit under the given conditions. It is precisely this question that can be calculated ex ante.

This fundamentally changes the nature of capital allocation. Decisions are not evaluated retrospectively, but systematically determined in advance. This reduces hidden opportunity costs, increases the quality of strategic allocation and turns investment management into a more precise, transparent and effective management tool.

For organizations with growing complexity, tight budgets and high capital commitments, this is not a methodological luxury. It is the logical next step in professional decision-making.

FAQ

What does ex ante mean in investment decisions?

Ex ante means that the optimal investment decision is calculated before implementation. It is not analyzed after the fact to determine whether a portfolio was good, but rather in advance to determine which combination generates the greatest benefit under the given restrictions.

Why is traditional prioritization not enough?

Because prioritization usually evaluates individual projects. However, the actual investment decision concerns combinations of projects. A good individual project is not automatically part of the best overall portfolio.

What is the difference between a local and a global optimum?

A local optimum is the best solution within a limited or pre-selected area. A global optimum is the best solution across all permissible combinations. Ex-ante calculation aims at the global optimum.

Why does the decision space grow exponentially?

Because every project basically has two states: It is implemented or not implemented. With N projects, this results in 2^N possible combinations. Even with medium portfolio sizes, this space becomes extremely large.

Why is a complete investment list so important?

Because only fully known options can be meaningfully optimized. If relevant projects or alternatives are missing, only a section of the real decision space is calculated. The global optimum can then remain invisible.

Is ex-ante calculation only relevant for large corporations?

No. It is relevant wherever several investment options are competing for limited funds at the same time. However, the benefit increases particularly strongly with larger portfolios, complex restrictions and high capital commitment.

What role does management still play then?

A central one. Management defines goals, restrictions, priorities and model logic. Mathematics does not replace management, but increases its precision and traceability.

Are scoring models or Monte Carlo simulations not enough?

They are useful tools, but they do not solve the optimization problem itself. Scorings evaluate projects, simulations analyze uncertainty. The question of the best combination of all permissible options can only be answered by optimization.

Why is this topic particularly important for fixed assets?

Because investments in fixed assets have a high capital commitment, long terms and low reversibility. Misallocations have a particularly long-lasting effect and are difficult to correct. This is why the quality of the ex-ante decision is particularly crucial.

What is the most important strategic benefit?

The most important benefit is better capital allocation. Ex-ante calculation reduces hidden opportunity costs and increases the likelihood that scarce resources will be used where they will generate the highest overall effect.