Zlatin Sarastov
PhD, Finance
UNWE, Sofia
MFin, London Business School

When analysis becomes cheap, judgement is what clients pay for.

I study and practise private banking, valuation under uncertainty and the economics of trust. Thirty-five years on both sides of the advisory table: building valuation models in Sofia in 1991, investment banking in London, running an airline, structuring investments for wealthy families, and now teaching and research.

Assistant Professor, Department of Finance, University of National and World Economy, Sofia.

Work

Private banking and AI

What a private bank still sells when analysis is automated: trust, discretion and judgement backed by something to lose.

Valuation under uncertainty

Stochastic discounted cash-flow models, Monte Carlo simulation and real options: valuing a business together with the value of its flexibility.

Advisory

Corporate finance, private placements and valuation for owners, families and investors, from the first model to the closing table.

Teaching

Banking, advanced bank management and private banking, in Bulgarian and English. Teaching finance at university level since 1993.

Book

The Banker and the Machine

Private banking in the age of artificial intelligence. The machine now does the analysis that advisers once sold; what remains scarce is trust, and a judgement someone is prepared to stand behind. The book draws on a research monograph, Swiss banks' flows after trust shocks, Eurosystem data on 84,755 households, the publicly announced uses of AI at 33 wealth managers, and three decades of practice.

In preparation. [Add publisher and date once agreed.]

Publications
  • 2026

    Private Banking in the Age of Artificial Intelligence Monograph

    Частното банкиране в ерата на изкуствения интелект. UNWE, Sofia. [status and publisher to confirm]

    Argues that the value of advice has two components: analysis, which AI is commoditising, and behavioural trust, which it cannot supply. Tests the claim on a coded review of 451 sources, client flows around trust shocks, a decade of private-bank results and household survey data.

  • 2017

    Valuing Investments under Uncertainty through Stochastic Modelling and Real Options PhD thesis

    Оценяване на инвестиции в условия на несигурност посредством стохастично моделиране и използване на реални опции. UNWE, Sofia. Supervisor: Emil Harsev.

    Builds a stochastic DCF in which sales growth and margins follow random processes, tests alternative distributions, and feeds the resulting volatility into real-options lattices. The closing argument: value the business as it stands, then add the value of management's flexibility.

  • 2012

    The Conte Marlo Method Article

    M. Dimov, M. Kamenarova and Z. Sarastov. Annual of the University of Mining and Geology "St. Ivan Rilski", vol. 55, part IV, pp. 68–73.

    A simplified Monte Carlo extension of the discounted cash-flow model that runs in an ordinary spreadsheet, shown on a hypothetical mining company. Born in a real negotiation, where a range of simulated futures settled a price that point estimates could not.

  • Ongoing

    Where new wealth is managed; what AI does for established banks Working papers

    Two research articles in preparation.

    The first follows new fortunes in China, India and the Gulf and where they choose to be managed. The second asks whether AI turns into productivity, and client money, for the incumbents in Zurich, London and the other established centres.

[Add further publications, conference papers and the ResearchGate link.]

Advisory

I advise owners, families and investors on decisions where the analysis is the easy part: selling or valuing a business, raising private capital, structuring cross-border investments and protecting creditors in a restructuring. Every engagement is confidential; I do not name clients.

  • 2007–

    Corporate finance and private placements

    Amfora Capital, Benefin Capital, Eurodaxx SA: disposals, capital raising, holding structures and investment in Central and Eastern Europe for wealthy clients in Europe's financial centres.

  • 2018–19

    Chief financial officer, tokenised share offering

    CFO of the company behind a regulated, tokenised share offering, with an investor roadshow across Asia and the Gulf. [Confirm naming and wording.]

  • 2005–07

    Chief executive, Bulgaria Air

    Led the national carrier through liberalisation and privatisation.

  • 2003–05

    Co-founder, Moneyport

    A UK money-transfer firm offering narrower spreads and faster transfers than the incumbents.

  • 2000–02

    Corporate finance, HSBC Investment Bank, London

    Technology origination team: valuing business models that had never existed before.

  • 1991–96

    Modeller, then partner

    Valuation models in Sofia from 1991; partner at 22 in a merchant finance house; board member of a new investment bank.

Models

Spreadsheet models for valuation under uncertainty

Working models that accompany my research and teaching, built to run in plain Excel without add-ins.

Conte Marlo: Monte Carlo DCF

A discounted cash-flow model in which sales growth follows a geometric Brownian motion and the operating margin reverts to its mean. Each row is one complete DCF; copied a few thousand times, the rows become a distribution of values rather than a single number.

Format
Excel workbook (.xlsx)
Method
GBM sales growth, Ornstein–Uhlenbeck margin, Monte Carlo
Reference
The Conte Marlo Method (2012)

Real options on a binomial lattice

Takes the volatility from the simulation and values the flexibility to expand, delay or abandon, using a Cox–Ross–Rubinstein lattice.

Format
Excel workbook (.xlsx)
Method
CRR binomial lattice, marketed asset disclaimer
Reference
PhD thesis (2017)
Contact

For research, speaking or advisory enquiries.

zlatin@sarastov.com